Valorisez vos réalisations phares sur People@HES-SO Plus d'infos
PEOPLE@HES-SO – Annuaire et Répertoire des compétences
PEOPLE@HES-SO – Annuaire et Répertoire des compétences

PEOPLE@HES-SO
Annuaire et Répertoire des compétences

Aide
language
  • fr
  • en
  • de
  • fr
  • en
  • de
  • SWITCH edu-ID
  • Administration
« Retour
Calvaresi Davide

Calvaresi Davide

Professeur-e HES Associé-e

Compétences principales

Multi-agent systems

Real-Time Multi-Agent Systems

Real-Time Systems

XAI

Explainable Multi-Agent Systems

Blockchain

Assistive technologies

  • Contact

  • Recherche

  • Publications

  • Conférences

Contrat principal

Professeur-e HES Associé-e

Bureau: ENP.23.N317

HES-SO Valais-Wallis - Haute Ecole d'Ingénierie
Rue de l'Industrie 23, 1950 Sion, CH
HEI - VS
Domaine
Technique et IT
Filière principale
Informatique et systèmes de communication
Aucune donnée à afficher pour cette section

En cours

EXPECTATION - Personalized Explainable Artificial Intelligence for decentralized agents with heterogeneous knowledge

Rôle: Requérant(e) principal(e)

Description du projet :

Explainable AI (XAI) has recently emerged proposing a set of techniques attempting to explain machine learning (ML) models. The recipients (explainee) are intended to be humans or other intelligent virtual entities. Transparency, trust, and debuging are the underlying features calling for XAI. However, in real-world settings, systems are distributed, data are heterogeneous, the "system" knowledge is bounded, and privacy concerns are subject to variable constraints. Current XAI approaches cannot cope with such requirements. Therefore, there is a need for personalized explainable artificial intelligence. We plan to develop models and mechanisms to reconcile sub-symbolic, symbolic, and semantic representations leveraging on the agent-based paradigm. In particular, the proposed approach combines inter-agent, intra-agent, and human-agent interactions to benefit from both the specialization of ML agents and the establishment of agent collaboration mechanisms, which will integrate heterogeneous knowledge / explanations extracted from efficient black-box AI agents. The project includes the validation of the personalization and heterogeneous knowledge integration approach through a prototype application in the domain of food and nutrition monitoring and recommendation, including the evaluation of agent-human explainability, and the performance of the employed techniques in a collaborative AI environment.

link: https://expectation.ehealth.hevs.ch/

Equipe de recherche au sein de la HES-SO: Calvaresi Davide

Partenaires académiques: Michael Schumacher, University of Applied Sciences and Arts Western Switzerland (HES-SO); Andrea Omicini, University of Bologna; Giovanni Ciatto, University of Bologna; Leon Van del Torre, University of Luxembourg; Amro Najjar, University of Luxembourg; Reyhan Aydogan, Ozyegin University

Durée du projet: 11.07.2021

Statut: En cours

Terminés

Axe Santé 2020 - HES-SO Valais-Wallis
AGP

Rôle: Collaborateur/trice

Financement: VS - Direction / Ra&D

Description du projet : Projet inter-disciplinaire pour l'Axe Santé de la HES-SO Valais-Wallis Principalement saisie d'heures pour la coordination et la participation aux séances de l'axe. Pas de remboursement de frais de déplacement inter-site. Ventilation des BSM : W00 - CC90014 - n° projet (ne pas mettre de pilier 6 dans la ventilation car c'est sur un centre de coûts) Répartition du budget sur les projets socle des instituts au 31.12.2020

Equipe de recherche au sein de la HES-SO: Prim Denis , Depeursinge Adrien , Bonazzi Riccardo , Loubier Jean-Christophe , Lugon Ralph , Weissbrodt Rafael , Verloo Henk , Antille Benoît , Zappellaz Nancy , Geiser Martial , Fournier Claude-Alexandre , Schumacher Michael Ignaz , Carrard Sophie , Bruchez Lise , Eggenschwiler Stefan , Caliesch Rahel , Soutrenon Mathieu , Aissaoui Djamel , Müller Henning , Hilfiker Roger , Solioz Emmanuel , Manzo Gaetano , Calbimonte Jean-Paul , Sattelmayer Martin , Simalatsar Alena , Giacomino Katia , Calvaresi Davide , Andrearczyk Vincent , Kaufmann Anne-Laure , Mastelic Joëlle , Segura Jean-Manuel , Bassolino Michela , Dini Sarah , Jovic Milica , Antille Alain , Dreyer-Cornut Sonam , Delaloye Matthieu , de Preux-Allet Lara , Voll Peter , Pfeifer Marc Emil , Cotting Alexandre , Eggel Ivan

Durée du projet: 01.01.2020 - 31.12.2020

Montant global du projet: 250'000 CHF

Statut: Terminé

Simulator of multi-agEnt systems enAbling tiMing reLiability and safEty critical analysiS of real-time and general-purpose algorithm
AGP

Rôle: Collaborateur/trice

Requérant(e)s: VS - Institut Informatique

Financement: Fondation Hasler

Description du projet : Context: In Cyber-Physical Systems (CPS), the correctness of the results does not solely depend on the correctness of the delivered value, it also depends on the time in which such a result is delivered. In the last decade, the Multi-Agent Systems (MAS) approach tackled distributed problem-solving, mainly focusing on optimal solutions and reasoning "about" but not "on" time (essential in safety-critical applications). -- Project Goals: MAS behavior is highly dynamical, therefore impossible to predict a priori. Hence, we aim at realizing a simulator supporting the in-depth analysis of MAS time reliability. In particular: (i) generating, simulating, and analyzing the MAS behaviors for general-purpose and safety-critical scenarios; (ii) providing detailed reports to study the timing-compliance of each MAS element. -- Expected Results: Time and system-related performance indexes will enable the analysis for identifying situations and factors compromising the system reliability (i.e., system compliance with strict timing constraints). Every simulated setup/scenario will be saved and made accessible for further investigation.

Equipe de recherche au sein de la HES-SO: Schumacher Michael Ignaz , Calvaresi Davide , Albanese Giuseppe

Partenaires académiques: VS - Institut Informatique

Durée du projet: 01.09.2019 - 31.08.2020

Montant global du projet: 49'985 CHF

Statut: Terminé

2024

Rethinking health recommender systems for active aging :
Article scientifique ArODES
an autonomy-based ethical analysis

Simona Tiribelli, Davide Calvaresi

Science and Engineering Ethics,  2024, vol. 30, no 22

Lien vers la publication

Résumé:

Health Recommender Systems are promising Articial-Intelligence-based tools endowing healthy lifestyles and therapy adherence in healthcare and medicine. Among the most supported areas, it is worth mentioning active aging. However, current HRS supporting AA raise ethical challenges that still need to be properly formalized and explored. This study proposes to rethink HRS for AA through an autonomy-based ethical analysis. In particular, a brief overview of the HRS’ technical aspects allows us to shed light on the ethical risks and challenges they might raise on individuals’ well-being as they age. Moreover, the study proposes a categorization, understanding, and possible preventive/mitigation actions for the elicited risks and challenges through rethinking the AI ethics core principle of autonomy. Finally, elaborating on autonomy-related ethical theories, the paper proposes an autonomy-based ethical framework and how it can foster the development of autonomy-enabling HRS for AA.

Towards interactive explanation-based nutrition virtual coaching systems
Article scientifique ArODES

Berk Buzcu, Melissa Tessa, Igor Tchappi, Amro Najjar, Joris Hulstijn, Davide Calvaresi, Reyhan Aydogan

Autonomous Agents and Multi-Agent Systems,  2024, 38, 5

Lien vers la publication

Résumé:

The awareness about healthy lifestyles is increasing, opening to personalized intelligent health coaching applications. A demand for more than mere suggestions and mechanistic interactions has driven attention to nutrition virtual coaching systems (NVC) as a bridge between human–machine interaction and recommender, informative, persuasive, and argumentation systems. NVC can rely on data-driven opaque mechanisms. Therefore, it is crucial to enable NVC to explain their doing (i.e., engaging the user in discussions (via arguments) about dietary solutions/alternatives). By doing so, transparency, user acceptance, and engagement are expected to be boosted. This study focuses on NVC agents generating personalized food recommendations based on user-specific factors such as allergies, eating habits, lifestyles, and ingredient preferences. In particular, we propose a user-agent negotiation process entailing run-time feedback mechanisms to react to both recommendations and related explanations. Lastly, the study presents the findings obtained by the experiments conducted with multi-background participants to evaluate the acceptability and effectiveness of the proposed system. The results indicate that most participants value the opportunity to provide feedback and receive explanations for recommendations. Additionally, the users are fond of receiving information tailored to their needs. Furthermore, our interactive recommendation system performed better than the corresponding traditional recommendation system in terms of effectiveness regarding the number of agreements and rounds.

2023

Exploring agent-based chatbots :
Article scientifique ArODES
a systematic literature review

Davide Calvaresi, Stefan Eggenschwiler, Yazan Mualla, Michael Schumacher, Jean-Paul Calbimonte

Journal of ambient intelligence and humanized computing,  2023, vol. 14, pp. 11207–11226

Lien vers la publication

Résumé:

In the last decade, conversational agents have been developed and adopted in several application domains, including education, healthcare, finance, and tourism. Nevertheless, chatbots still need to address several limitations and challenges, especially regarding personalization, limited knowledge-sharing capabilities, multi-domain campaign support, real-time monitoring, or integration of chatbot communities. To cope with these limitations, many approaches based on multi-agent systems models and technologies have been proposed in the literature, opening new research directions in this context. To better understand the current panorama of the different chatbot technology solutions employing agent-based methods, this Systematic Literature Review investigates the different application domains, end-users, requirements, objectives, technology readiness levels, designs, strengths, limitations, and future challenges of the solutions found in this scope. The results of this review are intended to provide researchers, software engineers, and innovators with a complete overview of the current state of the art and a discussion of the open challenges.

Characterization and future perspectives of virtual reality evacuation drills for safe built environments :
Article scientifique ArODES
a systematic literature review

Emanuele Gagliardi, Gabriele Bernardini, Enrico Quagliarini, Michael Schumacher, Davide Calvaresi

Safety Science,  July 2023, vol. 163, 106141

Lien vers la publication

Résumé:

Physical evacuation drills are pre-planned activities to train building occupants in facing emergencies and evaluate safety performances. Nowadays, technologies including Virtual Reality (VR) and Immersive Virtual Reality (IVR) are shifting from the physical to the virtual paradigm. AR enables just to extend real-world environment, while VR and IVR allow to (re)create and manipulate digital environments. VR and IVR simulation systems have been observed to guarantee higher involvement and long-term information retention — leveraging more attractive experiences and psychological arousal. However, efforts should be provided to improve end-user training while assessing occupants’ behaviors and the effectiveness of the emergency plan. This paper proposes a systematic literature review of VR and IVR evacuation solutions. To support and guide such effort, we formulated thirteen structured research questions investigating scenarios, recipients, requirements, objectives, methods, and technologies. The results mainly show that VR and IVR drills almost entirely tackle a single hazard, considers occupants as sole system recipients, and lack systems formalization. Among the most relevant outcomes, the paper analyzes the need for enhancing the modeling of emergency systems (e.g., signage, alarms), user inclusiveness (i.e., impaired individuals), devices, non-player characters, and additional effects (e.g., heat reproduction, sounds, and smells). These measures can improve the level of realism experienced by the user of IVR simulators and pave the way to more reliable outcomes.

Breast cancer survival analysis agents for clinical decision support
Article scientifique ArODES

Gaetano Manzo, Yves Pannatier, Patrick Duflot, Philippe Kolh, Marcela Chavez, Valérie Bleret, Davide Calvaresi, Oscar Jimenez-del-Toro, Michael Schumacher, Jean-Paul Calbimonte

Computer methods and programs in biomedicine,  Avril 2023, vol. 231

Lien vers la publication

Résumé:

Personalized support and assistance are essential for cancer survivors, given the physical and psychological consequences they have to suffer after all the treatments and conditions associated with this illness. Digital assistive technologies have proved to be effective in enhancing the quality of life of cancer survivors, for instance, through physical exercise monitoring and recommendation or emotional support and prediction. To maximize the efficacy of these techniques, it is challenging to develop accurate models of patient trajectories, which are typically fed with information acquired from retrospective datasets. This paper presents a Machine Learning-based survival model embedded in a clinical decision system architecture for predicting cancer survivors’ trajectories. The proposed architecture of the system, named PERSIST, integrates the enrichment and pre-processing of clinical datasets coming from different sources and the development of clinical decision support modules. Moreover, the model includes detecting high-risk markers, which have been evaluated in terms of performance using both a third-party dataset of breast cancer patients and a retrospective dataset collected in the context of the PERSIST clinical study.

Autonomous intersection management :
Article scientifique ArODES
optimal trajectories and efficient scheduling

Abdeljalil Abbas-Turki, Yazan Mualla, Nicolas Gaud, Davide Calvaresi, Wendan Du, Alexandre Lombard, Mahjoub Dridi, Abder Koukam

Sensors,  2023, vol. 23, no 3, p. 1509

Lien vers la publication

Résumé:

Intersections are at the core of congestion in urban areas. After the end of the Second World War, the problem of intersection management has benefited from a growing body of advances to address the optimization of the traffic lights’ phase splits, timing, and offset. These contributions have significantly improved traffic safety and efficiency in urban areas. However, with the growth of transportation demand and motorization, traffic lights show their limits. At the end of the 1990s, the perspective of autonomous and connected driving systems motivated researchers to introduce a paradigm shift for controlling intersections. This new paradigm is well known today as autonomous intersection management (AIM). It harnesses the self-organization ability of future vehicles to provide more accurate control approaches that use the smallest available time window to reach unprecedented traffic performances. This is achieved by optimizing two main points of the interaction of connected and autonomous vehicles at intersections: the motion control of vehicles and the schedule of their accesses. Considering the great potential of AIM and the complexity of the problem, the proposed approaches are very different, starting from various assumptions. With the increasing popularity of AIM, this paper provides readers with a comprehensive vision of noticeable advances toward enhancing traffic efficiency. It shows that it is possible to tailor vehicles’ speed and schedule according to the traffic demand by using distributed particle swarm optimization. Moreover, it brings the most relevant contributions in the light of traffic engineering, where flow–speed diagrams are used to measure the impact of the proposed optimizations. Finally, this paper presents the current challenging issues to be addressed.

2022

A DEXiRE for extracting propositional rules from neural networks via binarization
Article scientifique ArODES

Victor Contreras, Niccolo Marini, Lora Fanda, Gaetano Manzo, Yazan Mualla, Jean-Paul Calbimonte, Michael Schumacher, Davide Calvaresi

Electronics,  2022, vol. 11, no 24, p. 4171

Lien vers la publication

Résumé:

Background: Despite the advancement in eXplainable Artificial Intelligence, the expla nations provided by model-agnostic predictors still call for improvements (i.e., lack of accurate descriptions of predictors’ behaviors). Contribution: We present a tool for Deep Explanations and Rule Extraction (DEXiRE) to approximate rules for Deep Learning models with any number of hid den layers. Methodology: DEXiRE proposes the binarization of neural networks to induce Boolean functions in the hidden layers, generating as many intermediate rule sets. A rule set is inducted between the first hidden layer and the input layer. Finally, the complete rule set is obtained using inverse substitution on intermediate rule sets and first-layer rules. Statistical tests and satisfiability algorithms reduce the final rule set’s size and complexity (filtering redundant, inconsistent, and non-frequent rules). DEXiRE has been tested in binary and multiclass classifications with six datasets having different structures and models. Results: The performance is consistent (in terms of accuracy, fidelity, and rule length) with respect to the state-of-the-art rule extractors (i.e., ECLAIRE). Moreover, compared with ECLAIRE, DEXiRE has generated shorter rules (i.e., up to 74% fewer terms) and has shortened the execution time (improving up to 197% in the best-case scenario). Conclusions: DEXiRE can be applied for binary and multiclass classification of deep learning predictors with any number of hidden layers. Moreover, DEXiRE can identify the activation pattern per class and use it to reduce the search space for rule extractors (pruning irrelevant/redundant neurons)—shorter rules and execution times with respect to ECLAIRE.

Ethical and legal considerations for nutrition virtual coaches
Article scientifique ArODES

Davide Calvaresi, Rachele Carli, Jean-Gabriel Piguet, Victor H. Contreras, Gloria Luzzani, Amro Najjar, Jean-Paul Calbimonte, Michael Schumacher

AI and Ethics,  2022

Lien vers la publication

Résumé:

Choices and preferences of individuals are nowadays increasingly infuenced by countless inputs and recommendations provided by artifcial intelligence-based systems. The accuracy of recommender systems (RS) has achieved remarkable results in several domains, from infotainment to marketing and lifestyle. However, in sensitive use-cases, such as nutrition, there is a need for more complex dynamics and responsibilities beyond conventional RS frameworks. On one hand, virtual coaching systems (VCS) are intended to support and educate the users about food, integrating additional dimensions w.r.t. the conventional RS (i.e., leveraging persuasion techniques, argumentation, informative systems, and recommendation paradigms) and show promising results. On the other hand, as of today, VCS raise unexplored ethical and legal concerns. This paper discusses the need for a clear understanding of the ethical/legal-technological entanglements, formalizing 21 ethical and ten legal challenges and the related mitigation strategies. Moreover, it elaborates on nutrition sustainability as a further nutrition virtual coaches dimension for a better society.

A global taxonomy of interpretable AI :
Article scientifique ArODES
unifying the terminology for the technical and social sciences

Mara Graziani, Lidia Dutkiewicz, Davide Calvaresi, José Pereira Amorim, Katerina Yordanova, Mor Vered, Rahul Nair, Pedro Henriques Abreu, Tobias Blanke, Valeria Pulignano, John O. Prior, Lode Lauwaert, Wessel Reijers, Adrien Depeursinge, Vincent Andrearczyk, Henning Müller

Artificial intelligence review,  To be published

Lien vers la publication

Résumé:

Since its emergence in the 1960s, Artificial Intelligence (AI) has grown to conquer many technology products and their fields of application. Machine learning, as a major part of the current AI solutions, can learn from the data and through experience to reach high performance on various tasks. This growing success of AI algorithms has led to a need for interpretability to understand opaque models such as deep neural networks. Various requirements have been raised from different domains, together with numerous tools to debug, justify outcomes, and establish the safety, fairness and reliability of the models. This variety of tasks has led to inconsistencies in the terminology with, for instance, terms such as interpretable, explainable and transparent being often used interchangeably in methodology papers. These words, however, convey different meanings and are “weighted" differently across domains, for example in the technical and social sciences. In this paper, we propose an overarching terminology of interpretability of AI systems that can be referred to by the technical developers as much as by the social sciences community to pursue clarity and efficiency in the definition of regulations for ethical and reliable AI development. We show how our taxonomy and definition of interpretable AI differ from the ones in previous research and how they apply with high versatility to several domains and use cases, proposing a—highly needed—standard for the communication among interdisciplinary areas of AI.

2021

Cohort and trajectory analysis in multi-agent support systems for cancer survivors
Article scientifique ArODES

Gaetano Manzo, Davide Calvaresi, Oscar Alfonso Jiménez del Toro, Jean-Paul Calbimonte, Michael Schumacher

Journal of medical systems,  2021, vol. 45, article no. 109, pp. 1-10

Lien vers la publication

Résumé:

In the past decades, the incidence rate of cancer has steadily risen. Although advances in early and accurate detection have increased cancer survival chances, these patients must cope with physical and psychological sequelae. The lack of personalized support and assistance after discharge may lead to a rapid diminution of their physical abilities, cognitive impairment, and reduced quality of life. This paper proposes a personalized support system for cancer survivors based on a cohort and trajectory analysis (CTA) module integrated within an agent-based personalized chatbot named EREBOTS. The CTA module relies on survival estimation models, machine learning, and deep learning techniques. It provides clinicians with supporting evidence for choosing a personalized treatment, while allowing patients to benefit from tailored suggestions adapted to their conditions and trajectories. The development of the CTA within the EREBOTS framework enables to effectively evaluate the significance of prognostic variables, detect patient’s high-risk markers, and support treatment decisions.

Leveraging inter-tourists interactions via chatbots to bridge academia, tourism industries and future societies
Article scientifique ArODES

Davide Calvaresi, Ahmed Ibrahim, Jean-Paul Calbimonte, Emmanuel Fragnière, Roland Schegg, Michael Schumacher

Journal of tourism futures,  To be published

Lien vers la publication

Résumé:

Purpose : The tourism and hospitality sectors are experiencing radical innovation boosted by the advancements in Information and Communication Technologies. Increasingly sophisticated chatbots are introducing novel approaches, re-shaping the dynamics among tourists and service providers, and fostering a remarkable behavioral change in the overall sector. Therefore, the objective of this paper is two-folded: (1) to highlight the academic and industrial standing points with respect to the current chatbots designed/deployed in the tourism sector and (2) to develop a proof-of-concept embodying the most prominent opportunities in the tourism sector. Design/methodology/approach : This work elaborates on the outcomes of a Systematic Literature Review (SLR) and a Focus Group (FG) composed of experts from the tourism industry. Moreover, it presents a proof-of-concept relying on the outcomes obtained from both SLR and FG. Eventually, the proof-of-concept has been tested with experts and practitioners of the tourism sector. Findings : Among the findings elicited by this paper, we can mention the quick evolution of chatbot-based solutions, the need for continuous investments, upskilling, system innovation to tackle the eTourism challenges and the shift toward new dimensions (i.e. tourist-to-tourist-to-chatbot and personalized multi-stakeholder systems). In particular, we focus on the need for chatbot-based activity and thematic aggregation for next-generation tourists and service providers. Originality/value : Both academic- and industrial-centered findings have been structured and discussed to foster the practitioners' future research. Moreover, the proof-of-concept presented in the paper is the first of its kind, which raised considerable interest from both technical and business-planning perspectives.

The quest of parsimonious XAI :
Article scientifique ArODES
a human-agent architecture for explanation formulation

Yazan Mualla, Igor Tchappi, Timotheus Kampik, Amro Najjar, Davide Calvaresi, Abdeljalil Abbas-Turki, Stéphane Galland, Christophe Nicolle

Artificial intelligence,  2022, vol. 302, article no. 103573, pp. 1-26

Lien vers la publication

Résumé:

With the widespread use of Artificial Intelligence (AI), understanding the behavior of intelligent agents and robots is crucial to guarantee successful human-agent collaboration since it is not straightforward for humans to understand an agent's state of mind. Recent empirical studies have confirmed that explaining a system's behavior to human users fosters the latter's acceptance of the system. However, providing overwhelming or unnecessary information may also confuse the users and cause failure. For these reasons, parsimony has been outlined as one of the key features allowing successful human-agent interaction with parsimonious explanation defined as the simplest explanation (i.e. least complex) that describes the situation adequately (i.e. descriptive adequacy). While parsimony is receiving growing attention in the literature, most of the works are carried out on the conceptual front. This paper proposes a mechanism for parsimonious eXplainable AI (XAI). In particular, it introduces the process of explanation formulation and proposes HAExA, a human-agent explainability architecture allowing to make it operational for remote robots. To provide parsimonious explanations, HAExA relies on both contrastive explanations and explanation filtering. To evaluate the proposed architecture, several research hypotheses are investigated in an empirical user study that relies on well-established XAI metrics to estimate how trustworthy and satisfactory the explanations provided by HAExA are. The results are analyzed using parametric and non-parametric statistical testing.

One-to-many negotiation QoE management mechanism for end-user satisfaction
Article scientifique ArODES

Amro Najjar, Yazan Mualla, Kamal Deep Singh, Gauthier Picard, Davide Calvaresi

IEEE Access,  2021, vol. 9, pp. 59231-59243

Lien vers la publication

Résumé:

Quality of Experience (QoE) is defined as the measure of end-user satisfaction with the service. Most of the existing works addressing QoE-management rely on a binary vision of end-user satisfaction. This vision has been criticized by the growing empirical evidence showing that QoE is rather a degree. This article aims to go beyond the binary vision and propose a QoE management mechanism. We propose a one-to-many negotiation mechanism allowing the provider to undertake satisfaction management: to meet fine-grained user QoE goals, while still minimizing the costs. This problem is formulated as an optimization problem, for which a linear model is proposed. For reference, a generic linear program solver is used to find the optimal solution, and an alternative heuristic algorithm is devised to improve the responsiveness when the system has to scale up with a fast-growing number of users. Both are implemented and experimentally evaluated against state-of-the-art one-to-many negotiation frameworks.

EREBOTS :
Article scientifique ArODES
privacy-compliant agent-based platform for multi-scenario personalized health-assistant chatbots

Davide Calvaresi, Jean-Paul Calbimonte, Enrico Siboni, Stefan Eggenschwiler, Gaetano Manzo, Roger Hilfiker, Michael Schumacher

Electronics,  2021, vol. 10, no. 6, pp. article no. 666, pp. 1-30

Lien vers la publication

Résumé:

Context. Asynchronous messaging is increasingly used to support human–machine interactions, generally implemented through chatbots. Such virtual entities assist the users in activities of different kinds (e.g., work, leisure, and health-related) and are becoming ingrained into humans’ habits due to factors including (i) the availability of mobile devices such as smartphones and tablets, (ii) the increasingly engaging nature of chatbot interactions, (iii) the release of dedicated APIs from messaging platforms, and (iv) increasingly complex AI-based mechanisms to power the bots’ behaviors. Nevertheless, most of the modern chatbots rely on state machines (implementing conversational rules) and one-fits-all approaches, neglecting personalization, data-stream privacy management, multi-topic management/interconnection, and multimodal interactions. Objective. This work addresses the challenges above through an agent-based framework for chatbot development named EREBOTS. Methods. The foundations of the framework are based on the implementation of (i) multi-front-end connectors and interfaces (i.e., Telegram, dedicated App, and web interface), (ii) enabling the configuration of multi-scenario behaviors (i.e., preventive physical conditioning, smoking cessation, and support for breast-cancer survivors), (iii) online learning, (iv) personalized conversations and recommendations (i.e., mood boost, anti-craving persuasion, and balance-preserving physical exercises), and (v) responsive multi-device monitoring interface (i.e., doctor and admin). Results. EREBOTS has been tested in the context of physical balance preservation in social confinement times (due to the ongoing pandemic). Thirteen individuals characterized by diverse age, gender, and country distribution have actively participated in the experimentation, reporting advancements in the physical balance and overall satisfaction of the interaction and exercises’ variety they have been proposed.

Real-time multi-agent systems :
Article scientifique ArODES
rationality, formal model, and empirical results

Davide Calvaresi, Yashin Dicente Cid, Mauro Marinoni, Aldo Franco Dragoni, Amro Najjar, Michael Schumacher

Autonomous agents and multi-agent systems,  2021, vol. 35, article no. 12, pp. 1-37

Lien vers la publication

Résumé:

Since its dawn as a discipline, Artificial Intelligence (AI) has focused on mimicking the human mental processes. As AI applications matured, the interest for employing them into real-world complex systems (i.e., coupling AI with Cyber-Physical Systems—CPS) kept increasing. In the last decades, the multi-agent systems (MAS) paradigm has been among the most relevant approaches fostering the development of intelligent systems. In numerous scenarios, MAS boosted distributed autonomous reasoning and behaviors. However, many real-world applications (e.g., CPS) demand the respect of strict timing constraints. Unfortunately, current AI/MAS theories and applications only reason “about time” and are incapable of acting “in time” guaranteeing any timing predictability. This paper analyzes the MAS compliance with strict timing constraints (real-time compliance)—crucial for safety-critical applications such as healthcare, industry 4.0, and automotive. Moreover, it elicits the main reasons for the lack of real-time satisfiability in MAS (originated from current theories, standards, and implementations). In particular, traditional internal agent schedulers (general-purpose-like), communication middlewares, and negotiation protocols have been identified as co-factors inhibiting real-time compliance. To pave the road towards reliable and predictable MAS, this paper postulates a formal definition and mathematical model of real-time multi-agent systems (RT-MAS). Furthermore, this paper presents the results obtained by testing the dynamics characterizing the RT-MAS model within the simulator MAXIM-GPRT. Thus, it has been possible to analyze the deadline miss ratio between the algorithms employed in the most popular frameworks and the proposed ones. Finally, discussing the obtained results, the ongoing and future steps are outlined.

2020

Dynamic consent management for clinical trials via private blockchain technology
Article scientifique ArODES

Giuseppe Albanese, Jean-Paul Calbimonte, Michael Schumacher, Davide Calvaresi

Journal of ambient intelligence and humanized computing,  2020, vol. 11, no. 11, pp. 4909–4926

Lien vers la publication

Résumé:

Clinical trials (CTs) are essential for the advancement of medical research, paving the way for the development and adoption of new treatments, and contributing to the evolution of healthcare. An essential factor for the success of a CT is the appropriate management of its participants and their personal data. According to the current regulations, collecting and using personal data from participants must comply with rigorous standards. Therefore, healthcare institutes need to obtain freely given, specific, informed, and unambiguous consent before being able to collect the data. Some of the major limitations of the current technological solutions are the lack of control over the granularity of consent grants, as well as the difficulty of handling dynamic changes of consent over time. In this paper, we present SCoDES, an approach for trusted and decentralized management of dynamic consent in clinical trials, based on blockchain technology (BCT). The usage of blockchain provides a set of features that allow maintaining consent information with trust guarantees while avoiding the need for a dedicated or centralized third trusted party. We provide a full implementation of SCoDES, made available as a self-contained infrastructure, with the possibility to interact with external services, and using hyperledger as a blockchain framework.

Agent-based modeling for ontology-driven analysis of patient trajectories
Article scientifique ArODES

Davide Calvaresi, Michael Schumacher, Jean-Paul Calbimonte

Journal of medical systems,  2020, vol. 44, no. 9, article 158

Lien vers la publication

Résumé:

Patients are often required to follow a medical treatment after discharge, e.g., for a chronic condition, rehabilitation after surgery, or for cancer survivor therapies. The need to adapt to new lifestyles, medication, and treatment routines, can produce an individual burden to the patient, who is often at home without the full support of healthcare professionals. Although technological solutions –in the form of mobile apps and wearables– have been proposed to mitigate these issues, it is essential to consider individual characteristics, preferences, and the context of a patient in order to offer personalized and effective support. The specific events and circumstances linked to an individual profile can be abstracted as a patient trajectory, which can contribute to a better understanding of the patient, her needs, and the most appropriate personalized support. Although patient trajectories have been studied for different illnesses and conditions, it remains challenging to effectively use them as the basis for data analytics methodologies in decentralized eHealth systems. In this work, we present a novel approach based on the multi-agent paradigm, considering patient trajectories as the cornerstone of a methodology for modelling eHealth support systems. In this design, semantic representations of individual treatment pathways are used in order to exchange patient-relevant information, potentially fed to AI systems for prediction and classification tasks. This paper describes the major challenges in this scope, as well as the design principles of the proposed agent-based architecture, including an example of its use through a case scenario for cancer survivors support.

Real-time compliant stream processing agents for physical rehabilitation
Article scientifique ArODES

Davide Calvaresi, Jean-Paul Calbimonte

Sensors,  2020, vol. 20, no. 3, pp. 1-34

Lien vers la publication

Résumé:

Digital rehabilitation is a novel concept that integrates state-of-the-art technologies for motion sensing and monitoring, with personalized patient-centric methodologies emerging from the field of physiotherapy. Thanks to the advances in wearable and portable sensing technologies, it is possible to provide patients with accurate monitoring devices, which simplifies the tracking of performance and effectiveness of physical exercises and treatments. Employing these approaches in everyday practice has enormous potential. Besides facilitating and improving the quality of care provided by physiotherapists, the usage of these technologies also promotes the personalization of treatments, thanks to data analytics and patient profiling (e.g., performance and behavior). However, achieving such goals implies tackling both technical and methodological challenges. In particular, (i) the capability of undertaking autonomous behaviors must comply with strict real-time constraints (e.g., scheduling, communication, and negotiation), (ii) plug-and-play sensors must seamlessly manage data and functional heterogeneity, and finally (iii) multi-device coordination must enable flexible and scalable sensor interactions. Beyond traditional top-down and best-effort solutions, unsuitable for safety-critical scenarios, we propose a novel approach for decentralized real-time compliant semantic agents. In particular, these agents can autonomously coordinate with each other, schedule sensing and data delivery tasks (complying with strict real-time constraints), while relying on ontology-based models to cope with data heterogeneity. Moreover, we present a model that represents sensors as autonomous agents able to schedule tasks and ensure interactions and negotiations compliant with strict timing constraints. Furthermore, to show the feasibility of the proposal, we present a practical study on upper and lower-limb digital rehabilitation scenarios, simulated on the MAXIM-GPRT environment for real-time compliance. Finally, we conduct an extensive evaluation of the implementation of the stream processing multi-agent architecture, which relies on existing RDF stream processing engines.

2019

The good, the bad, and the ethical implications of bridging blockchain and multi-agent systems
Article scientifique ArODES

Davide Calvaresi, Jean-Paul Calbimonte, Alevtina Dubovitskaya, Valerio Mattioli, Jean-Gabriel Piguet, Michael Schumacher

Information,  2019, vol. 10, no. 12, article 363

Lien vers la publication

Résumé:

The agent based approach is a well established methodology to model distributed intelligent systems. Multi-Agent Systems (MAS) are increasingly employed in applications dealing with safety and information critical tasks (e.g., in eHealth, financial, and energy domains). Therefore, transparency and the trustworthiness of the agents and their behaviors must be enforced. For example, employing reputation based mechanisms can promote the development of trust. Nevertheless, besides recent early stage studies, the existing methods and systems are still unable to guarantee the desired accountability and transparency adequately. In line with the recent trends, we advocate that combining blockchain technology (BCT) and MAS can achieve the distribution of the trust, removing the need for trusted third parties (TTP), potential single points of failure. This paper elaborates on the notions of trust, BCT, MAS, and their integration. Furthermore, to attain a trusted environment, this manuscript details the design and implementation of a system reconciling MAS (based on the Java Agent DEvelopment Framework (JADE)) and BTC (based on Hyperledger Fabric). In particular, the agents’ interactions, computation, tracking the reputation, and possible policies for disagreement-management are implemented via smart contracts and stored on an immutable distributed ledger. The results obtained by the presented system and similar solutions are also discussed. Finally, ethical implications (i.e., opportunities and challenges) are elaborated before concluding the paper.

Towards profile and domain modelling in agent-based applications for behavior change
Chapitre de livre ArODES

Jean-Paul Calbimonte, Davide Calvaresi, Fabien Dubosson, Michael Schumacher

Dans Demazeau, Yves, Advances in practical applications of survivable agents and multi-agent systems : the PAAMS collection  (pp. 16-28). 2019,  Cham : Springer

Lien vers la publication

Résumé:

Health support programs play a vital role in public health and prevention strategies at local and national levels, for issues such as smoking cessation, physical rehabilitation, nutrition, or to regain mobility. A key success factor in these topics is related to the appropriate use of behavior change techniques, as well as tailored recommendations for users/patients, adapted to their goals and the continuous monitoring of their progress. Social networks interactions and the use of multi-agent technologies can further improve the effectiveness of these programs, especially through personalization and profiling of users and patients. In this paper we propose an agent-based model for supporting behavior change in eHealth programs. Moreover, we identify the main challenges in this area, especially regarding profile and domain modeling profiles for healthcare behavioral programs, where the definition of goals, expectations and argumentation play a key role in the success of a intervention.

A startup assessment approach based on multi-agent and blockchain technologies
Chapitre de livre ArODES

Davide Calvaresi, Ekaterina Voronova, Jean-Paul Calbimonte, Valerio Mattioli, Michael Schumacher

Dans De La Prieta. Fernando, et al., Highlights of practical applications of survivable agents and multi-agent systems. The PAAMS collection: International Workshops of PAAMS 2019, Ávila, Spain, June 26–28, 2019, Proceedings  (12 p.). 2019,  Cham : Springer

Lien vers la publication

Résumé:

The dynamic nature of startups is linked to both high risks in investments as well as potentially important financial benefits. A key aspect to manage interactions among investors, experts, and startups, is the establishment of trust guarantees. This paper presents the formalization and implementation of a system enforcing trust in the startup assessment domain. To do so, an existing architecture has been extended, incorporating a multi-agent community and related interactions via private blockchain technology. The developed system enables a trust-based community, immutably storing, tracking, and monitoring the agents’ interactions and reputations.

Real-time multi-agent systems for telerehabilitation scenarios
Article scientifique ArODES

Davide Calvaresi, Mauro Marinoni, Aldo Franco Dragoni, Roger Hilfiker, Michael Schumacher

Artificial intelligence in medicine,  2019, vol. 96, pp. 217-231

Lien vers la publication

Résumé:

Telerehabilitation in older adults is most needed in the patient environments, rather than in formal ambulatories or hospitals. Supporting such practices brings significant advantages to patients, their family, formal and informal caregivers, clinicians, and researchers. This paper presents a focus group with experts in physiotherapy and telerehabilitation, debating on the requirements, current techniques and technologies developed to facilitate and enhance the effectiveness of telerehabilitation, and the still open challenges. Particular emphasis is given to (i) the body-parts requiring the most rehabilitation, (ii) the typical environments, initial causes, and general conditions, (iii) the values and parameters to be observed, (iv) common errors and limitations of current practices and technological solutions, and (v) the envisioned and desired technological support. Consequently, it has been performed a systematic review of the state of the art, investigating what types of systems and support currently cope with telerehabilitation practices and possible matches with the outcomes of the focus group. Technological solutions based on video analysis, wearable devices, robotic support, distributed sensing, and gamified telerehabilitation are examined. Particular emphasis is given to solutions implementing agent-based approaches, analyzing and discussing strength, limitations, and future challenges. By doing so, it has been possible to relate functional requirements expressed by professional physiotherapists and researchers, with the need for extending multi-agent systems (MAS) peculiarities at the sensing level in wearable solutions establishing new research challenges. In particular, to be employed in safety-critical cyber-physical scenarios with user-sensor and sensor-sensor interactions, MAS are requested to handle timing constraints, scarcity of resources and new communication means, crucial to providing real-time feedback and coaching. Therefore, MAS pillars such as the negotiation protocol and the agent's internal scheduler have been investigated, proposing solutions to achieve the aforementioned real-time compliance.

2018

Trust in tourism via blockchain technology :
Chapitre de livre ArODES
results from a systematic review

Davide Calvaresi, Maxine Leis, Alevtina Dubovitskaya, Roland Schegg, Michael Schumacher

Dans Neidhardt, Julia, Pesonen, Juho, Information and Communication Technologies in Tourism 2019 : proceedings of the International Conference in Nicosia, Cyprus, January 30–February 1, 2019  (12 p.). 2018,  Cham : Springer

Lien vers la publication

Résumé:

Trust-free and trust-regulated systems based on blockchain technology (BCT) are currently experiencing the maximum hype and promise to revolutionise entire domains. Tourism products (intangible services) are highly dependent on trust and reputation management that is traditionally centralised and delegated to “expected” reliable third-parties (e.g., TripAdvisor). Although BCT has only recently started approaching the tourism industry and being employed in real-world applications, the scientific community has already been extensively exploring the promises of BCT. Therefore, there is an impending need for organising and understanding current knowledge and formalise societal, scientific, and technological challenges of applying BCT in the tourism industry. This paper moves the first step, presenting a systematic scientific literature review of studies involving BCT for tourism purposes. Providing a comprehensive overview, actors, assumptions, requirements, strengths, and limitations characterising the state of the art are analysed. Finally, advantages and future challenges of applying BCT in the tourism area are discussed.

Multi-agent systems and blockchain :
Chapitre de livre ArODES
results from a systematic literature review

Davide Calvaresi, Alevtina Dubovitskaya, Jean-Paul Calbimonte, Kuldar Taveter, Michael Schumacher

Dans An, Bo, Bajo, Javier, Demazeau, Yves, Fernández-Caballero, Antonio, Advances in practical applications of agents, multi-agent systems, and complexity : the PAAMS collection : 16th International Conference, PAAMS 2018, Toledo, Spain, June 20–22, 2018, Proceedings  (Pp. 110-126). 2018,  Cham : Springer

Lien vers la publication

Résumé:

Multi-Agent Systems (MAS) technology is widely used for the development of intelligent distributed systems that manage sensitive data (e.g., ambient assisted living, healthcare, energy trading). To foster accountability and trusted interactions, recent trends advocate the use of blockchain technologies (BCT) for MAS. Although most of these approaches have only started exploring the topic, there is an impending need for establishing a research road-map, as well as identifying scientific and technological challenges in this scope. As a first necessary step towards this goal, this paper presents a systematic literature review of studies involving MAS and BCT as reconciling solutions. Aiming at providing a comprehensive overview of their application domains, we analyze motivations, assumptions, requirements, strengths, and limitations presented in the current state of the art. Moreover, discussing the future challenges, we introduce our vision on how MAS and BCT could be combined in different application scenarios.

MAXIM-GPRT :
Chapitre de livre ArODES
a simulator of local schedulers, negotiations, and communication for multi-agent systems in general-purpose and real-time scenarios

Giuseppe Albanese, Davide Calvaresi, Paolo Sernani, Fabien Dubosson, Aldo Franco Dragoni, Michael Schumacher

Dans An, Bo, Bajo, Javier, Demazeau, Yves, Fernández-Caballero, Antonio, Advances in practical applications of agents, multi-agent systems, and complexity : The PAAMS collection : 16th International Conference, PAAMS 2018, Toledo, Spain, June 20–22, 2018, Proceedings  (Pp. 291-295). 2018,  Cham : Springer

Lien vers la publication

Résumé:

In safety-critical scenarios, the compliance with strict-timing constraints is mandatory. This demo presents a simulator named MAXIM-GPRT enabling the analysis of the behaviors produced by Multi-Agent Systems (MAS) composed of both General-Purpose (GP) and Real-time (RT) algorithms. Therefore, MAXIM-GPRT is crucial to prove that current MAS cannot provide timing guarantees, nor guarantee correct behaviors in the worst case scenario. However, adopting and adapting models and algorithms from RT systems, such a compliance, can be achieved.

Multi-agent systems' negotiation protocols for cyber-physical systems :
Chapitre de livre ArODES
results from a systematic literature review

Davide Calvaresi, Kevin Appoggetti, Luca Lustrissimini, Mauro Marinori, Paolo Sernani, Aldo Franco Dragoni, Michael Schumacher

Dans Rocha, Ana Paula, Proceedings of the 10th International Conference on Agents and Artificial Intelligence  (pp. 224-235). 2018,  Setubal : SciTePress

Lien vers la publication

Résumé:

Cyber Physical Systems (CPS) require a multitude of components interacting among themselves and with the users to perform automatic actions, usually under unpredictable or uncertain conditions. Multi-Agent Systems (MAS) have emerged over the years as one of the major technological paradigms regulating interactions and negotiations among autonomous entities running under heterogeneous conditions. As such, MAS have the potential to support CPS in implementing a highly reconfigurable distributed thinking. However, some gaps are still present between MAS’ features and the strict requirements of CPS. The most relevant is the lack of reliability, which is mainly due to specific features characterizing negotiation protocols. This paper presents a systematic literature review of MAS negotiation protocols aiming at providing a comprehensive overview of their strengths and limitations, examining both the assumptions and requirements set during their development. While this work confirms the potential of MAS in regulating the interactions among CPS components, the findings also highlight the absence of real-time compliance in current negotiation protocols. Strongly characterizing CPS, the capability to face strict time constraints could bridge the gap between MAS and CPS.

2024

Evaluation of traffic controller performance via systematic exploration
Conférence ArODES

Krešimir Kušic, Davide Calvaresi, Amy Liffey, Lora Fanda, Martin Greguric, Edouard Ivanjko, René Schumann

2024 International Symposium ELMAR

Lien vers la conférence

Résumé:

Traffic controllers must operate reliably across diverse traffic states. Due to the stochastic non-linear characteristics of traffic flow, commonly used feedback-based controllers require parameter tuning for each specific traffic regime, which is done offline using simulations. Generating representative traffic scenarios for large-scale simulations is often computationally expensive. To reduce the computational burden, this paper proposes a systematic exploration of the Structured Simulation Framework (SSF). This approach aims to approximate controller performance with a minimal number of simulations, by adjusting the parameter space continuously to regions where controller performances are weakly approximated. This process continues until controller performance is well approximated across the entire input domain. Results show SSF convergence of performance estimate of the controller while reducing the number of required simulations. This helps identify traffic scenarios where the controller performs poorly, and, thus, can be used as a framework towards guided controller tuning.

Evaluation of the user-centric explanation strategies for interactive recommenders
Conférence ArODES

Berk Buzcu, Emre Kuru, Davide Calvaresi, Reyhan Aydogan

Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS 2024)

Lien vers la conférence

Résumé:

As recommendation systems become increasingly prevalent in numerous fields, the need for clear and persuasive interactions with users is rising. Integrating explainability into these systems is emerging as an effective approach to enhance user trust and sociability. This research focuses on recommendation systems that utilize a range of explainability techniques to foster trust by providing understandable personalized explanations for the recommendations made. In line with this, we study three distinct explanation methods that correspond with three basic recommendation strategies and assess their efficacy through user experiments. The findings from the experiments indicate that the majority of participants value the suggested explanation styles and favor straightforward, concise explanations over comparative ones.

The wildcard XAI :
Conférence ArODES
from a necessity, to a resource, to a dangerous decoy

Rachele Carli, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS 2024)

Lien vers la conférence

Résumé:

There has been a growing interest in Explainable Artificial Intelligence (henceforth XAI) models among researchers and AI programmers in recent years. Indeed, the development of highly interactive technologies that can collaborate closely with users has made explainability a necessity. This intends to reduce mistrust and the sense of unpredictability that AI can create, especially among non-experts. Moreover, the potential of XAI as a valuable resource has been recognized, considering that it can make intelligent systems more user-friendly and reduce the negative impact of black box systems. Building on such considerations, the paper discusses the potential dangers of large language models (LLMs) that generate explanations to support the outcomes produced. While these models may give users the illusion of control over the system’s responses, they actually have persuasive and non-explanatory effects. Therefore, it is argued here that XAI, appropriately regulated, should be a resource to empower users of AI systems. Any other apparent explanations should be reported to avoid misleading and circumventing effects.

Explanation of deep learning models via logic rules enhanced by embeddings analysis, and probabilistic models
Conférence ArODES

Victor Contreras, Michael Schumacher, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS 2024)

Lien vers la conférence

Résumé:

Deep Learning (DL) models are increasingly dealing with heterogeneous data (i.e., a mix of structured and unstructured data), calling for adequate eXplainable Artificial Intelligence (XAI) methods. Nevertheless, only some of the existing techniques consider the uncertainty inherent to the data. To this end, this study proposes a pipeline to explain heterogeneous data-based DL models by combining embedding analysis, rule extraction methods, and probabilistic models. The proposed pipeline has been tested using synthetic data (multi-individual food items tracking). This study has achieved (i) inference enhancement through probabilistic and evidential reasoning, (ii) generation of logical explanations based on extracted rules and predictions, and (iii) integration of textual data into the explanation pipeline through embedding analysis.

A framework for explainable multi-purpose virtual assistants :
Conférence ArODES
a nutrition-focused case study

Berk Buzcu, Yvan Pannatier, Reyhan Aydogan, Michael Schumacher, Jean-Paul Calbimonte, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS 2024)

Lien vers la conférence

Résumé:

Existing agent-based chatbot frameworks need seamless mechanisms to include explainable dialogic engines within the contextual flow. To this end, this paper presents a set of novel modules within the EREBOTS agent-based framework for chatbot development, including dialog-based plug-and-play custom algorithms, agnostic back/front ends, and embedded interactive explainable engines that can manage human feedback at run time. The framework has been employed to implement an explainable agent-based interactive food recommender system. The latter has been tested with 44 participants, who followed a nutrition recommendation interaction series, generating explained recommendations and suggestions, which were, in general, well received. Additionally, the participants provided important insights to be included in future work.

Towards dynamic self-organizing wearables for head and neck digital rehabilitation
Conférence ArODES

Berk Buzcu, Davide Calvaresi, Banani Anuraj, Jean-Paul Calbimonte

DEBS '24: Proceedings of the 18th ACM International Conference on Distributed and Event-based Systems

Lien vers la conférence

Résumé:

Digital rehabilitation is dramatically changing the way in which physiotherapists conduct their practice and analyse the exercises of patients. As opposed to traditional treatment with episodic verification of the therapist, patients can perform prescribed exercises at home supported by personalised assistive technologies and wearable devices. This work presents a prototype that highlights the integration of motion data streams from wearable sensors in the context of head and neck rehabilitation exercises. The system consists of self-organising devices placed in shoulders, neck, and head, set up following low-code interaction flows. Patients can interact with the platform through a tablet App that provides feedback through real-time 3D avatars and tracks data for post-exercise analytics.

2023

Study-Buddy :
Conférence ArODES
a knowledge graph-powered learning companion for school students

Fernanda Martinez, Diego Collarana, Davide Calvaresi, Martin Arispe, Carla Florida, Jean-Paul Calbimonte

The Semantic Web: ESWC 2023 Satellite Events

Lien vers la conférence

Résumé:

Large Language Models (LLMs) have the potential to substantially improve educational tools for students. However, they face limitations, including factual accuracy, personalization, and the lack of control over the sources of information. This paper presents Study-Buddy, a prototype of a conversational AI assistant for school students to address the above-mentioned limitations. Study-Buddy embodies an AI assistant based on a knowledge graph, LLMs models, and computational persuasion. It is designed to support educational campaigns as a hybrid AI solution. The demonstrator showcases interactions with Study-Buddy and the crucial role of the Knowledge Graph for the bot to present the appropriate activities to the students. A video demonstrating the main features of Study-Buddy is available at: https://youtu.be/DHPTsN1RI9o.

Explanation generation via decompositional rules extraction for head and neck cancer classification
Conférence ArODES

Victor Contreras, Andrea Bagante, Niccolò Marini, Michael Schumacher, Vincent Andrearczyk, Davide Calvaresi

Explainable and transparent AI and multi-agent systems

Lien vers la conférence

Résumé:

Human papillomavirus (HPV) accounts for 60% of head and neck (H&N) cancer cases. Assessing the tumor extension (tumor grading) and determining whether the tumor is caused by HPV infection (HPV status) is essential to select the appropriate treatment. Therefore, developing non-invasive, transparent (trustworthy), and reliable methods is imperative to tailor the treatment to patients based on their status. Some studies have tried to use radiomics features extracted from positron emission tomography (PET) and computed tomography (CT) images to predict HPV status. However, to the best of our knowledge, no research has been conducted to explain (e.g., via rule sets) the internal decision process executed on deep learning (DL) predictors applied to HPV status prediction and tumor grading tasks. This study employs a decompositional rule extractor (namely DEXiRE) to extract explanations in the form of rule sets from DL predictors applied to H&N cancer diagnosis.The extracted rules can facilitate researchers’ and clinicians’ understanding of the model’s decisions (making them more transparent) and can serve as a base to produce semantic and more human-understandable explanations.

Reinterpreting vulnerability to tackle deception in principles-based XAI for human-computer interaction
Conférence ArODES

Rachele Carli, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems

Lien vers la conférence

Résumé:

Artificial intelligence (AI) systems have been increasingly adopted for decision support, behavioral change purposes, assistance, and aid in daily activities and decisions. Thus, focusing on design and interaction that, in addition to being functional, foster users’ acceptance and trust is increasingly necessary. Human-computer interaction (HCI) and human-robot interaction (HRI) studies focused more and more on the exploitation of communication means and interfaces to possibly enact deception. Despite the literal meaning often attributed to the term, deception does not always denote a merely manipulative intent. The expression “banal deception” has been theorized to specifically refer to design strategies that aim to facilitate the interaction. Advances in explainable AI (XAI) could serve as technical means to minimize the risk of distortive effects on people’s perceptions and will. However, this paper argues that how the provided explanations and their content can exacerbate the deceptive dynamics or even manipulate the end user. Therefore, in order to avoid similar consequences, this analysis suggests legal principles to which the explanation must conform to mitigate the side effects of deception in HCI/HRI. Such principles will be made enforceable by assessing the impact of deception on the end users based on the concept of vulnerability – understood here as the rationalization of the inviolable right of human dignity – and control measures implemented in the given systems.

2022

Human-social robots interaction :
Conférence ArODES
the blurred line between necessary anthropomorphization and manipulation

Rachele Carli, Amro Najjar, Davide Calvaresi

HAI '22: Proceedings of the 10th International Conference on Human-Agent Interaction

Lien vers la conférence

Résumé:

In the context of human-social robot interaction, it has been proventhat an affable design and the ability to exhibit emotional andsocial skills are central to fostering acceptance and more efficientsystem performance. Nevertheless, these features may result inmanipulative dynamics, able to impact the psychological sphere ofthe users, affecting their ability to make decisions and to exercisefree, conscious will. This highlights the need to identify a legalframework that balances the interests at stake. To this end, theprinciple of human dignity is proposed here as a criterion to ensure(i) the protection of users’ fundamental rights, and (ii) an effectiveand truly human-friendly technological development

GB-Flex :
Conférence ArODES
automated and distributed decision-making in energy balancing groups

Davide Calvaresi, Khoa Nguyen, René Schumann

Abstracts of the 11th DACH+ Conference on Energy Informatics

Lien vers la conférence

Résumé:

This paper briefly summarizes the possible strategies and shows the feasibility and profits of automatizing the BGs.

Risk and exposure of XAI in persuasion and argumentation :
Conférence ArODES
the case of manipulation

Rachele Carli, Amro Najjar, Davide Calvaresi

Explainable and transparent AI and multi-agent systems : 4th International Workshop, EXTRAAMAS 2022

Lien vers la conférence

Résumé:

In the last decades, Artificial intelligence (AI) systems have been increasingly adopted in assistive (possibly collaborative) decision-making tools. In particular, AI-based persuasive technologies are designed to steer/influence users’ behaviour, habits, and choices to facilitate the achievement of their own – predetermined – goals. Nowadays, the inputs received by the assistive systems leverage heavily AI data-driven approaches. Thus, it is imperative to have transparent and understandable (to the user) both the process leading to the recommendations and the recommendations. The Explainable AI (XAI) community has progressively contributed to “opening the black box”, ensuring the interaction’s effectiveness, and pursuing the safety of the individuals involved. However, principles and methods ensuring the efficacy and information retain on the human have not been introduced yet. The risk is to underestimate the context dependency and subjectivity of the explanations’ understanding, interpretation, and relevance. Moreover, even a plausible (and possibly expected) explanation can lead to an imprecise or incorrect outcome or its understanding. This can lead to unbalanced and unfair circumstances, such as giving a financial advantage to the system owner/provider and the detriment of the user. This paper highlights that the sole explanations – especially in the context of persuasive technologies – are not self-sufficient to protect users’ psychological and physical integrity. Conversely, explanations could be misused, becoming themselves a tool of manipulation. Therefore, we suggest characteristics safeguarding the explanation from being manipulative and legal principles to be used as criteria for evaluating the operation of XAI systems, both from an ex-ante and ex-post perspective.

Integration of local and global features explanation with global rules extraction and generation tools
Conférence ArODES

Victor Contreras, Michael Schumacher, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems : 4th International Workshop, EXTRAAMAS 2022, Virtual Event, May 9–10, 2022, Revised Selected Papers

Lien vers la conférence

Résumé:

Widely used in a growing number of domains, Deep Learning predictors are achieving remarkable results. However, the lack of transparency (i.e., opacity) of their inner mechanisms has raised trust and employability concerns. Nevertheless, several approaches fostering models of interpretability and explainability have been developed in the last decade. This paper combines approaches for local feature explanation (i.e., Contextual Importance and Utility – CIU) and global feature explanation (i.e., Explainable Layers) with a rule extraction system, namely ECLAIRE. The proposed pipeline has been tested in four scenarios employing a breast cancer diagnosis dataset. The results show improvements such as the production of more human-interpretable rules and adherence of the produced rules with the original model.

A personalized agent-based chatbot for nutritional coaching
Conférence ArODES

Davide Calvaresi, Stefan Eggenschwiler, Jean-Paul Calbimonte, Gaetano Manzo, Michael Schumacher

Proceedings of WI-IAT '21: IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology

Lien vers la conférence

Résumé:

Intelligent systems increasingly support users’ behavior change, including exercise adherence, smoking cessation, and healthy diet adoption. Their effectiveness is affected by the personalization degree of advice/coaching and HMI mechanisms. This paper proposes a personalized agent-based chatbot platform assisting the user in healthy nutrition via pervasive technologies leveraging dynamical, multi-modal, and personalized interactions. The system provides diet recommendations and tracks the user’s food intake and nutritional behaviors to promote a healthy lifestyle. The study concludes with a user study and performance evaluation.

2021

Study of context-based personalized recommendations for points of interest
Conférence ArODES

Gaetano Manzo, Davide Calvaresi, Jean-Paul Calbimonte, Okoro Esteem, Michael Schumacher

Proceedings of the 2021 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT 2021)

Lien vers la conférence

Résumé:

Location-based services are essential to delivering information for users in the context of travel, leisure, and sports application. Nevertheless, these services are often implemented as recommendations and suggestions that may overwhelm users, or fail to adapt to their goals, behavior, and context. To address these limitations, this paper presents NearMe, an application that provides tailored recommendations of Points of Interest surrounding the user. Beyond existing approaches, NearMeallows the generation of dynamic recommendations from heterogeneous service providers, and the definition of regions to which notifications are related. Moreover, it allows to fine-tune notifications, thus preventing over-information and noise. A preliminary study has been conducted involving a heterogeneous group of potential users and service providers that elaborates on their vision, expectations, features desiderata, and possible interfaces.

On explainable negotiations via argumentation
Conférence ArODES

Victor Contreras, Reyhan Aydogan, Amro Najjar, Davide Calvaresi

Proceedings of BNAIC/BeneLearn 2021: 33rd Benelux Conference on Artificial Intelligence and 30th Belgian-Dutch Conference on Machine Learning

Lien vers la conférence

Expectation :
Conférence ArODES
personalized explainable artificial intelligence for decentralized agents with heterogeneous knowledge

Davide Calvaresi, Giovanni Ciatto, Amro Najjar, Reyhan Aydogan, Leon Van der Torre, Andrea Omicini, Michael Schumacher

Explainable and Transparent AI and Multi-Agent Systems : Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3–7, 2021, Revised Selected Papers

Lien vers la conférence

Résumé:

Explainable AI (XAI) has emerged in recent years as a set of techniques and methodologies to interpret and explain machine learning (ML) predictors. To date, many initiatives have been proposed. Nevertheless, current research efforts mainly focus on methods tailored to specific ML tasks and algorithms, such as image classification and sentiment analysis. However, explanation techniques are still embryotic, and they mainly target ML experts rather than heterogeneous end-users. Furthermore, existing solutions assume data to be centralised, homogeneous, and fully/continuously accessible—circumstances seldom found altogether in practice. Arguably, a system-wide perspective is currently missing. The project named “Personalized Explainable Artificial Intelligence for Decentralized Agents with Heterogeneous Knowledge ” (Expectation) aims at overcoming such limitations. This manuscript presents the overall objectives and approach of the Expectation project, focusing on the theoretical and practical advance of the state of the art of XAI towards the construction of personalised explanations in spite of decentralisation and heterogeneity of knowledge, agents, and explainees (both humans or virtual). To tackle the challenges posed by personalisation, decentralisation, and heterogeneity, the project fruitfully combines abstractions, methods, and approaches from the multi-agent systems, knowledge extraction/injection, negotiation, argumentation, and symbolic reasoning communities.

To pay or not to pay attention :
Conférence ArODES
classifying and interpreting visual selective attention frequency features

Laura Fanda, Yashin Dicente Cid, Pawel J. Matusz, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems : Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3–7, 2021, Revised Selected Papers

Lien vers la conférence

Résumé:

Selective attention is the ability to promote the processing of objects important for the accomplishment of our behavioral goals (target objects) over the objects not important to those goals (distractor objects). Previous investigations have shown that the mechanisms of selective attention contribute to enhancing perception in both simple daily tasks and more complex activities requiring learning new information. Recently, it has been verified that selective attention to target objects and distractor objects is separable in the frequency domain, using Logistic Regression (LR) and Support Vector Machines (SVMs) classification. However, discerning dynamics of target and distractor objects in the context of selective attention has not been accomplished yet. This paper extends the investigations on the possible classification and interpretation of distraction and intention solely relying on neural activity (frequency features). In particular, this paper (i) classifies distractor objects vs. target object replicating the LR classification of prior studies, extending the analysis by (ii) interpreting the coefficient weights relating to all features with a focus on N2PC features, and (iii) retrains an LR classifier with the features deemed important by the interpretation analysis. As a result of the interpretation methods, we have successfully decreased the feature size to 7.3% of total features – i.e., from 19,072 to 1,386 features – while recording only a 0.04 loss in performance accuracy score—i.e., from 0.65 to 0.61. Additionally, the interpretation of the classifiers’ coefficient weights unveiled new evidence regarding frequency which has been discussed along with the paper.

Towards explainable visionary agents :
Conférence ArODES
license to dare and imagine

Giovanni Ciatto, Amro Najjar, Jean-Paul Calbimonte, Davide Calvaresi

Explainable and Transparent AI and Multi-Agent Systems : Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3–7, 2021, Revised Selected Papers

Lien vers la conférence

Résumé:

Since their appearance, computer programs have embodied discipline and structured approaches and methodologies. Yet, to this day, equipping machines with imaginative and creative capabilities remains one of the most challenging and fascinating goals we pursue. Intelligent software agents can behave intelligently in well-defined scenarios, relying on Machine Learning (ML), symbolic reasoning, and the ability of their developers for tailoring smart behaviors to specific application domains. However, to forecast the evolution of all possible scenarios is unfeasible. Thus, intelligent agents should autonomously/creatively adapt to the world’s mutability. This paper investigates the meaning of imagination in the context of cognitive agents. In particular, it addresses techniques and approaches to let agents autonomously imagine/simulate their course of action and generate explanations supporting it, and formalizes thematic challenges. Accordingly, we investigate research areas including: (i) reasoning and automatic theorem proving to synthesize novel knowledge via inference; (ii) automatic planning and simulation, used to speculate over alternative courses of action; (iii) machine learning and data mining, exploited to induce new knowledge from experience; and (iv) biochemical coordination, which keeps imagination dynamic by continuously reorganizing it.

Explainable and transparent artificial intelligence and multi-agent systems :
Conférence ArODES
third International workshop, EXTRAAMAS 2021, virtual event, May 3–7, 2021, revised selected papers

Davide Calvaresi, Amro Najjar, Michael Winikoff, Kary Främling

EXTRAAMAS: International workshop on explainable, transparent autonomous agents and multi-agent systems

Lien vers la conférence

Résumé:

This book constitutes the proceedings of the Third International Workshop on Explainable, Transparent AI and Multi-Agent Systems, EXTRAAMAS 2021, which was held virtually due to the COVID-19 pandemic. The 19 long revised papers and 1 short contribution were carefully selected from 32 submissions. The papers are organized in the following topical sections: XAI & machine learning; XAI vision, understanding, deployment and evaluation; XAI applications; XAI logic and argumentation; decentralized and heterogeneous XAI.

Orchestrating tourism actors' network via the "N-1 N+1 touchpoints" algorithm :
Conférence ArODES
a b2b chatbot to improve customer's journeys

Randolf Ramseyer, Davide Calvaresi, Benjamin Nanchen, Roland Schegg, Michael Schumacher, Emmanuel Fragnière

Proceedings of the 19th International Conference e-Society 2021

Lien vers la conférence

Résumé:

This paper elaborates on a novel concept to orchestrate tourism networks. In particular, each actor optimizes his service (touchpoint n) while seamlessly transferring the customer information from the previous (touchpoint n-1) to the following service (touchpoint n+1). To implement such a theory, we leveraged on chatbot technology, which interfacing directly with the user eased the transition from the point of interaction (touchpoint) n-1 to n+1. Moreover, the chatbot entails the connection between the nodes of the customer's journey, enabling user profiling, personalization, and knowledge transfer. The deployment of a chatbot implementing the "n-1 n+1 touchpoints" model would significantly benefit actors operating in a fragmented touristic economy (i.e., Switzerland). Hence, we tested the first prototype in the heart of the Canton Valais, where, in collaboration with students in tourism, we based the scenario on counterfactual thinking. In particular, we identified all the possible situations a grandmother and her grandson might face arriving at the Sierre train station to spend a day in Crans-Montana. In turn, using reenactment theatre techniques, tourism professionals played the worst-case scenario (without chatbot) and the best case (with the chatbot) to elicit the different clients' perceptions on those diametral situations. Such a feasibility study paved the way to a more holistic view, employing artificial intelligence techniques to enhance the chatbot and smoothen the "n-1 n+1 touchpoints" dynamics.

A platform for difficulty assessment and recommendation of hiking trails
Conférence ArODES

Jean-Paul Calbimonte, Simon Martin, Davide Calvaresi, Alexandre Cotting

Information and communication technologies in tourism 2021 : proceedings of the ENTER 2021 eTourism Conference

Lien vers la conférence

Résumé:

In recent years, the popularity of hiking has steadily increased across different segments of the population. Although there is considerable evidence of the benefits for hikers regarding physical and mental health, the inherent risks of these outdoor activities cannot be underestimated. Accident prevention and an increase of awareness about possible risks are necessary to minimize hiking and pedestrian tourism’s negative consequences. In most hiking information maps and interactive applications, there is usually not enough information about difficulty points or the granularity level required to provide tailored recommendations to hikers with physical or psychological limitations. In this paper, we present Syris, a geo-information system for hiking itineraries that incorporates Points-Of-Difficulty to assess the level of effort, technique, and risk of hiking trails. The system allows users to filter itineraries and obtain recommendations based on the assessment of difficulty following a well-established methodology. The system has been implemented, deployed and tested with real data in the region of Val d’Anniviers in Switzerland, and is openly available to enable further developments and refinement.

The evolution of chatbots in tourism :
Conférence ArODES
a systematic literature review

Davide Calvaresi, Ahmed Ibrahim, Jean-Paul Calbimonte, Roland Schegg, Emmanuel Fragnière, Michael Schumacher

Information and Communication Technologies in Tourism 2021 : Proceedings of the ENTER 2021 eTourism Conference, January 19–22, 2021

Lien vers la conférence

Résumé:

In the last decade, Information and Communication Technologies have revolutionized the tourism and hospitality sector. One of the latest innovations shaping new dynamics and fostering a remarkable behavioral change in the interaction between the service provider and the tourist is the employment of increasingly sophisticated chatbots. This work analyzes the most recent systems presented in the literature (since 2016) investigated via 12 research questions. The often appreciated quick evolution of such solutions is the primary outcome. However, such technological and financial fast-pace requires continuous investments, upskilling, and system innovation to tackle the eTourism challenges, which are shifting towards new dimensions.

2020

Decentralized management of patient profilesand trajectories through semantic web agents
Conférence ArODES

Jean-Paul Calbimonte, Davide Calvaresi, Michael Schumacher

Proceedings of the 3rd International Workshop on Semantic Web Meets Health Data Management (SWH 2020)

Lien vers la conférence

Résumé:

The usage of healthcare data for analytics and patient applications has increased in recent years opening a number of technical, ethical and scientific challenges. Among these, those related to the management of personal and sensitive health data have been addressed through decentralized solutions for patient data, often implemented and modelled using distributed agents and semantic technologies. In this paper, we present a technical summary of our previous works in this area, comprising efforts to: (i) use ontology models to represent patient trajectories,(ii) employ agent-based architectures to model and employ decentralized patient data exchanges, (iii) define agent cooperation and negotiation strategies for healthcare data interactions, (iv) adopt semantic data models for privacy-aware agents, and (v) implement multi-agent systems for real-time healthcare data processing

Ethical concerns and opportunities in binding intelligent systems and blockchain technology
Conférence ArODES

Davide Calvaresi, Jean-Gabriel Piguet, Jean-Paul Calbimonte, Timotheus Kampik, Amro Najjar, Guillaume Gadek, Michael Schumacher

Proceedings of International Workshops of PAAMS 2020 : Highlights in Practical Applications of Agents, Multi-Agent Systems, and Trust-worthiness.

Lien vers la conférence

Résumé:

Intelligent systems are becoming increasingly complex and pervade a broad range of application domains, including safety-critical systems such as e-health, finance, and energy management. Traditional approaches are no longer capable of addressing the demand for trust and transparency in these applications. Hence, the current decade is demanding intelligent systems to be autonomous, and in particular explainable, transparent, and trustworthy. To satisfy such requirements, and therefore to comply with the recent EU regulations in the matter (e.g., GDPR), intelligent systems (e.g., Multi-Agent Systems - MAS) and technologies enabling tamper-proof and distributed consensus (e.g., Blockchain Technology - BCT) are conveying into reconciling solutions. Recently, the empowerment of MAS with BCT (and the use of BCT themselves) has gained considerable momentum, raising challenges, and unveiling opportunities. However, several ethical concerns have yet to be faced. This paper elaborates on the entanglement among ethical and technological challenges while proposing and discussing approaches that address these emerging research opportunities.

Personal data privacy semantics in multi-agent systems interactions
Conférence ArODES

Davide Calvaresi, Michael Schumacher, Jean-Paul Calbimonte

Proceedings of the 18th International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS 2020) : Advances in Practical Applications of Agents, Multi-Agent Systems, and Trustworthiness

Lien vers la conférence

Résumé:

In recent years, we have witnessed the growth of applications relying on the use and processing of personal data, especially in the health and well-being domains. Users themselves produce these data (e.g., through self-reported data acquisition, or personal devices such as smartphones, smartwatches or other wearables). A key challenge in this context is to guarantee the protection of personal data privacy, respecting the rights of users for deciding about data reuse, consent to data processing and storage, anonymity conditions, or the right to withhold or delete personal data. With the enforcement of recent regulations in this domain, such as the GDPR, applications are required to guarantee compliance, challenging current practices for personal data management. In this paper, we address this problem in the context of decentralized personal data applications, which may need to interact and negotiate conditions of data processing and reuse. Following a distributed paradigm without a top-down organization, we propose an agent-based model in which personal data providers and data consumers are embedded into privacy-aware agents capable of negotiating and coordinating data reuse, consent, and policies, using semantic vocabularies for privacy and provenance.

RT-BDI :
Conférence ArODES
a real-time BDI model

Francesco Alzetta, Paolo Giorgini, Mauro Marinoni, Davide Calvaresi

Proceedings of the 18th International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS 2020) : Advances in Practical Applications of Agents, Multi-Agent Systems, and Trustworthiness

Lien vers la conférence

Résumé:

Currently, distributed cyber-physical systems (CPS) rely upon embedded real-time systems, which can guarantee compliance with time constraints. CPS are increasingly required to act and interact with one another in dynamic environments. In the last decades, the Belief-Desire-Intention (BDI) architecture has proven to be ideal for developing agents with flexible behavior. However, current BDI models can only reason about time and not in time. This lack prevents BDI agents from being adopted in designing CPS, and particularly in safety-critical applications. This paper proposes a revision of the BDI model by integrating real-time mechanisms into the reasoning cycle of the agent. By doing so, the BDI agent can make decisions and execute plans ensuring compliance with strict timing constraints also in dynamic environments, where unpredictable events may occur.

SEAMLESS :
Conférence ArODES
simulation and analysis for multi-agent system in time-constrained environments

Davide Calvaresi, Giuseppe Albanese, Jean-Paul Calbimonte, Michael Schumacher

Proceedings of the 18th International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS 2020) : Advances in Practical Applications of Agents, Multi-Agent Systems, and Trustworthiness

Lien vers la conférence

Résumé:

The correctness of a system operating in time-constrained scenarios leverages on both precision and delivery time of its outcome. This paper presents SEAMLESS, a system enabling the design, simulation, and in-depth analysis of Multi-Agent Systems (MAS). In particular, SEAMLESS allows defining in detail the agents’ knowledge (set of tasks it might execute), needs (set of tasks to be negotiated), local scheduler (execution of the task-set), negotiation protocols, possible communication delays, and heuristics related to the parameters mentioned above. This tool is pivotal in the strive to study and realize real-time MAS.

Agent-Based Explanations in AI :
Conférence ArODES
towards an abstract framework

Giovanni Ciatto, Michael Schumacher, Andrea Omicini, Davide Calvaresi

Proceedings of the EXTRAAMAS: International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems 2020

Lien vers la conférence

Résumé:

Recently, the eXplainable AI (XAI) research community has focused on developing methods making Machine Learning (ML) predictors more interpretable and explainable. Unfortunately, researchers are struggling to converge towards an unambiguous definition of notions such as interpretation, or, explanation—which are often (and mistakenly) used interchangeably. Furthermore, despite the sound metaphors that Multi-Agent System (MAS) could easily provide to address such a challenge, and agent-oriented perspective on the topic is still missing. Thus, this paper proposes an abstract and formal framework for XAI-based MAS, reconciling notions, and results from the literature.

In-time explainability in multi-agent systems :
Conférence ArODES
challenges, opportunities, and roadmap

Francesco Alzetta, Paolo Giorgini, Amro Najjar, Michael Schumacher, Davide Calvaresi

Proceedings of the EXTRAAMAS: International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems 2020

Lien vers la conférence

Résumé:

In the race for automation, distributed systems are required to perform increasingly complex reasoning to deal with dynamic tasks, often not controlled by humans. On the one hand, systems dealing with strict-timing constraints in safety-critical applications mainly focused on predictability, leaving little room for complex planning and decision-making processes. Indeed, real-time techniques are very efficient in predetermined, constrained, and controlled scenarios. Nevertheless, they lack the necessary flexibility to operate in evolving settings, where the software needs to adapt to the changes of the environment. On the other hand, Intelligent Systems (IS) increasingly adopted Machine Learning (ML) techniques (e.g., subsymbolic predictors such as Neural Networks). The seminal application of those IS started in zero-risk domains producing revolutionary results. However, the ever-increasing exploitation of ML-based approaches generated opaque systems, which are nowadays no longer socially acceptable—calling for eXplainable AI (XAI). Such a problem is exacerbated when IS tend to approach safety-critical scenarios. This paper highlights the need for on-time explainability. In particular, it proposes to embrace the Real-Time Beliefs Desires Intentions (RT-BDI) framework as an enabler of eXplainable Multi-Agent Systems (XMAS) in time-critical XAI.

An abstract framework for agent-based explanations in AI
Conférence ArODES

Giovanni Ciatto, Davide Calvaresi, Michael Schumacher, Andrea Omicini

Proceedings of the International Conference on Autonomous Agents and Multi-Agent Systems 2020

Lien vers la conférence

Résumé:

We propose an abstract framework for XAI based on MAS encompassing the main definitions and results from the literature, focussing on the key notions of interpretation and explanation.

Semantic data models for hiking trail difficulty assessment
Conférence ArODES

Jean-Paul Calbimonte, Simon Martin, Davide Calvaresi, Nancy Zappelaz, Alexandre Cotting

Proceedings of Information and Communication Technologies in Tourism 2020 : proceedings of the International Conference in Surrey

Lien vers la conférence

Résumé:

Hiking is a popular outdoor activity that if practiced regularly, can bring significant health benefits. Moreover, considering that hikers range from expert mountaineers to older adults with limited physical capabilities, it touches a large target audience, and is strategically included in several tourism packages across the globe. Thus, a precise characterization of the tracks, especially regarding their points of difficulty, is crucial to effectively cope with the challenge of identifying the best-suited hiking trails for heterogeneous users. This paper introduces a semantic model for representing and integrating the main characteristics of a track, including their different types of difficulties, using Semantic Web ontologies. The construction of knowledge graphs that use such a model may constitute a first step towards a system for personalized recommendations of trails based on difficulty-classification criteria.

2019

Towards XMAS :
Conférence ArODES
eXplainable and trustworthy Multi-Agent Systems

Giovanni Ciatto, Roberta Calegari, Andrea Omicini, Davide Calvaresi

Proceedings of the Proceedings of the 1st Workshop on Artificial Intelligence and Internet of Things co-located with the 18th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2019)

Lien vers la conférence

Résumé:

In the context of the Internet of Things (IoT), intelligent systems (IS) are increasingly relying on Machine Learning (ML) techniques. Given the opaqueness of most ML techniques, however, humans have to rely on their intuition to fully understand the IS outcomes: helping them is the target of eXplainable Arti_cial Intelligence (XAI). Current solutions { mostly too speci_c, and simply aimed at making ML easier to interpret { cannot satisfy the needs of IoT, characterised by heterogeneous stimuli, devices, and data-types concurring in the composition of complex information structures. Moreover, Multi-Agent Systems (MAS) achievements and advancements are most often ignored, even when they could bring about key features like explainability and trustworthiness. Accordingly, in this paper we (i) elicit and discuss the most signi_cant issues a_ecting modern IS, and (ii) devise the main elements and related interconnections paving the way towards reconciling interpretable and explainable IS using MAS.

Social network chatbots for smoking cessation :
Conférence ArODES
agent and multi-agent frameworks

Davide Calvaresi, Jean-Paul Calbimonte, Fabien Dubosson, Michael Schumacher, Amro Najjar

Proceedings of WI '19 : IEEE/WIC/ACM International Conference on Web Intelligence

Lien vers la conférence

Résumé:

Asynchronous messaging is leading human-machine interaction due to the boom of mobile devices and social networks. The recent release of dedicated APIs from messaging platforms boosted the development of computer programs able to conduct conversations, (i.e., chatbots), which have been adopted in several domain-specific contexts. This paper proposes SMAG: a chatbot framework supporting a smoking cessation program (JDF) deployed on a social network. In particular, it details the single-agent implementation, the campaign results, a multi-agent design for SMAG enabling the modelization of personalized behavior and user profiling, and highlighting of coupling chatbot technology with and multi-agent systems.

Explainable agents and robots :
Conférence ArODES
results from a systematic literature review

Sule Anjomshoae, Najjar Amro, Davide Calvaresi, Kary Främling

Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS) 2019

Lien vers la conférence

Résumé:

Humans are increasingly relying on complex systems that heavily adopts Artificial Intelligence (AI) techniques. Such systems are employed in a growing number of domains, and making them explainable is an impelling priority. Recently, the domain of eXplainable Artificial Intelligence (XAI) emerged with the aims of fostering transparency and trustworthiness. Several reviews have been conducted. Nevertheless, most of them deal with data-driven XAI to overcome the opaqueness of black-box algorithms. Contributions addressing goal-driven XAI (e.g., explainable agency for robots and agents) are still missing. This paper aims at filling this gap, proposing a Systematic Literature Review. The main findings are (i) a considerable portion of the papers propose conceptual studies, or lack evaluations or tackle relatively simple scenarios; (ii) almost all of the studied papers deal with robots/agents explaining their behaviors to the human users, and very few works addressed inter-robot (inter-agent) explainability. Finally, (iii) while providing explanations to non-expert users has been outlined as a necessity, only a few works addressed the issues of personalization and context-awareness.

Stream reasoning agents
Conférence ArODES

Riccardo Tommasini, Davide Calvaresi, Jean-Paul Calbimonte

Proceedings of the International Conference on Autonomous Agent and Multi-Agent Systems (AAMAS)

Lien vers la conférence

Résumé:

Data streams are increasingly needed for different types of applications and domains, where dynamicity and data velocity are of foremost importance. In this context, research challenges raise regarding the generation, publication, processing, and discovery of these streams, especially in distributed, heterogeneous and collaborative environments such as the Web. Stream reasoning has addressed some of these challenges in the last decade, presenting a novel data processing paradigm that lays at the intersection among semantic data modeling, stream processing, and inference techniques. However, stream reasoning works have focused almost exclusively on architectures and approaches that assume an isolated processing environment. Therefore, they lack, in general, the means for discovering, collaborating, negotiating, sharing, or validating data streams on a highly heterogeneous ecosystem as the Web. Agents and multi-agent systems research has long developed principles and foundations for enabling some of these features, although usually under assumptions that require to be revised in order to comply with the characteristics of data streams. This paper presents a vision for a Web of stream reasoning agents, capable of sharing not only streaming data, but also processing duties, using collaboration and negotiation protocols, while relying on common vocabularies and protocols that take into account the high dynamicity of their knowledge, goals, and behavioral patterns.

Explainable multi-agent systems through blockchain technology
Conférence ArODES

Davide Calvaresi, Yazan Mualla, Amro Najjar, Stéphane Galland, Michael Schumacher

Proceedings of the EXTRAAMAS 2019 International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems

Lien vers la conférence

Résumé:

Advances in Artificial Intelligence (AI) are contributing to a broad set of domains. In particular, Multi-Agent Systems (MAS) are increasingly approaching critical areas such as medicine, autonomous vehicles, criminal justice, and financial markets. Such a trend is producing a growing AI-Human society entanglement. Thus, several concerns are raised around user acceptance of AI agents. Trust issues, mainly due to their lack of explainability, are the most relevant. In recent decades, the priority has been pursuing the optimal performance at the expenses of the interpretability. It led to remarkable achievements in fields such as computer vision, natural language processing, and decision-making systems. However, the crucial questions driven by the social reluctance to accept AI-based decisions may lead to entirely new dynamics and technologies fostering explainability, authenticity, and user-centricity. This paper proposes a joint approach employing both blockchain technology (BCT) and explainability in the decision-making process of MAS. By doing so, current opaque decision-making processes can be made more transparent and secure and thereby trustworthy from the human user standpoint. Moreover, several case studies involving Unmanned Aerial Vehicles (UAV) are discussed. Finally, the paper discusses roles, balance, and trade-offs between explainability and BCT in trust-dependent systems.

2018

Trusted registration, negotiation, and service evaluation in multi-agent systems throughout the Blockchain Technology
Conférence ArODES

Davide Calvaresi, Alevtina Dubovitskaya, Diego Retaggi, Aldo F. Dragoni, Michael Schumacher

Proceedings of the 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI)

Lien vers la conférence

Résumé:

Some recent trends in distributed intelligent systems rely extensively on agent-based approaches. The so-called Multi- Agent Systems (MAS) are taking over the management of sensitive data on behalf of their producers and users (e.g., medical records, financial investment, energy market). Therefore, trusted interactions are needed more than ever, while accountability and transparency among the agents seem crucial characteristics to be achieved. To do so, recent trends advocate the use of blockchain technologies (BCT) in MAS. The blockchain is a distributed ledger technology that can execute programmable transaction logic, and provides a shared, immutable, and transparent appendonly register of all the actions happening in the network. Although a few theoretical approaches have already been proposed, the quest for such a system consolidating BCT and MAS to guarantee privacy, scalability, transparency, and efficiency continues. This paper presents a reconciling system including BCT within the dynamics of a MAS. Such a system aims at (i) building a solid ground for trusted interactions and (ii) enabling more characterizing feature-based and trustworthy ways of computing agent reputation. The system has been tested in four scenarios with different configurations (regular executions and involving down-agents or malicious behaviors). Finally, the paper summarizes and discusses the experience gained, argues about the strategic choice of binding MAS and BCT, and presents some future challenges.

Timing reliability for local schedulers in multi-agent systems
Conférence ArODES

Davide Calvaresi, Giuseppe Albanese, Mauro Marinoni, Fabien Dubosson, Paolo Sernani, Aldo Franco Dragoni, Michael Schumacher

Proceedings of the 1st International Workshop on Real-Time compliant Multi-Agent Systems co-located with the Federated Artificial Intelligence Meeting

Lien vers la conférence

Résumé:

In the last decades, the use of Multi-Agent Systems (MAS) resulted in being the most relevant approach to foster the development of systems performing distributed reasoning, automated/autonomous actions, and regulating component interactions in unpredictable and uncertain scenarios. The scientific community provided numerous innovative contributions about resource and task allocation seeking for optimal/sub-optimal solutions. The adoption of MAS in Cyber- Physical Systems (CPS) is producing outstanding results. However, in current MAS, the actual task execution is still delegated to traditional general-purpose scheduling algorithms running within the agent (local scheduler of behaviors). The main consequence is the incapability to enforce compliance with strict timing constraints (i.e., the impossibility of providing any guarantee about the system’s behavior in the worst-case scenario). Therefore, the adoption of MAS is hampered, excluding significant application scenarios such as safety-critical environments. This paper proposes the schedulability analysis of various task-sets, that are feasible using real-time schedulers, on top of traditional general-purpose solutions. In particular, the study of deadline-missing rate occurring in general-purpose setups, evaluated on an agent-based simulator developed on OMNET++, named MAXIM-GPRT, is presented. The obtained results strengthen the motivations for adopting and adapting real-time scheduling mechanisms as the local scheduler within agents.

A task-sets generator for supporting the analysis of multi-agent systems under general purpose and real-time conditions
Conférence ArODES

Davide Calvaresi, Giuseppe Albanese, Fabien Dubosson, Mauro Marinoni, Michael Schumacher

Proceedings of the 1st International Workshop on Real-Time compliant Multi-Agent Systems co-located with the Federated Artificial Intelligence Meeting

Lien vers la conférence

Résumé:

The adoption of Multi-Agent Systems (MAS) is permeating Internet of Things (IoT) and Cyber-Physical Systems (CPS). Timing reliability of MAS is a daring challenge. The study of local task execution and negotiation of workloads are catalyzing considerable interest. By adopting techniques typical of Real-Times Systems (RTS), MAS’s ability to comply with strict timing constraints has been proven. However, a complete formalization is still missing, and some of the existing mathematical models introduce considerable pessimism in the performance analysis. Therefore, the need for tools supporting the study of the behavior of agent-based systems is rising. Particularly, the capability of systematic assessment and comparison of their performance. This paper presents a system to generate task-sets and operating scenarios, to support the study of timing reliability, behavior, and performance of MAS. The parameters required for such a generation are characterized by randomly extracted values (e.g., the number of agents, single agent utilization factors, and single task utilization factor). For each parameter, it is possible to select a given statistical distribution to be applied to user-defined ranges. In particular, logic, constraints, and dependencies characterizing the generation algorithm are detailed and framed in a functional work-flow. Moreover, such a system integrates a MAS simulator powered by both general-purpose and real-time algorithms, named MAXIMGPRT. Hence, the presented tool is also able to show the logs of the tested scenarios equipped with graphs to enable the performance analysis.

2017

Local scheduling in multi-agent systems :
Conférence ArODES
getting ready for safety-critical scenarios

Davide Calvaresi, Mauro Marinoni, Luca Lustrissimini, Kevin Appoggetti, Paolo Sernani, Michael Schumacher, Giorgio Buttazzo

Proceedings of 15th European Conference on Multi-Agent Systems (EUMAS 2017)

Lien vers la conférence

Résumé:

Multi-Agent Systems (MAS) have been supporting the development of distributed systems performing decentralized thinking and reasoning, automated actions, and regulating component interactions in unpredictable and uncertain scenarios. Despite the scientific literature is plenty of innovative contributions about resource and tasks allocation, the agents still schedule their behaviors and tasks by employing traditional general-purpose scheduling algorithms. By doing so, MAS are unable to enforce the compliance with strict timing constraints. Thus, it is not possible to provide any guarantee about the system behavior in the worst-case scenario. Thereby, as they are, they cannot operate in safety-critical environments. This paper analyzes the agents' local schedulers provided by the most relevant agent-based frameworks from a cyber-physical systems point of view. Moreover, it maps a set of agents' behaviors on task models from the real-time literature. Finally, a practical case-study is provided to highlight how such "MAS reliability" can be achieved.

The challenge of real-time multi-agent systems for enabling IoT and CPS
Conférence ArODES

Davide Calvaresi, Mauro Marinoni, Arnon Sturm, Michael Schumacher, Giorgio Buttazzo

Proceedings of the International Conference on Web Intelligence (WI '17)

Lien vers la conférence

Résumé:

Techniques originating from the Internet of Things (IoT) and Cyber-Physical Systems (CPS) areas have extensively been applied to develop intelligent and pervasive systems such as assistive monitoring, feedback in telerehabilitation, energy management, and negotiation. Those application do-mains particularly include three major characteristics: intel-ligence, autonomy and real-time behavior. Multi-Agent Sys-tems (MAS) are one of the major technological paradigms that are used to implement such systems. However, they mainly address the first two characteristics, but miss to com-ply with strict timing constraints. The timing compliance is crucial for safety-critical applications operating in domains such as healthcare and automotive. The main reasons for this lack of real-time satisfiability in MAS originate from cur-rent theories, standards, and technological implementations. In particular, internal agent schedulers, communication mid-dlewares, and negotiation protocols have been identified as co-factors inhibiting the real-time compliance. This paper provides an analysis of such MAS components and pave the road for achieving the MAS compliance with strict timing constraints, thus fostering reliability and predictability.

Agent-based systems for telerehabilitation :
Conférence ArODES
strengths, limitations and future challenges

Davide Calvaresi, Michael Schumacher, Mauro Marinoni, Roger Hil?ker, Aldo F. Dragoni, Giorgio Buttazzo

Proceedings of X Workshop on Agents Applied in Health Care, A2HC 2017

Lien vers la conférence

Résumé:

Telerehabilitation in older adults is most needed in the patient environments, rather than in formal ambulatories or hospitals. Sup-porting such practices brings significant advantages to patients, their family, formal and informal caregivers, clinicians, and researchers. Sev-eral techniques and technologies have been developed aiming at facilitat-ing and enhancing the effectiveness of telerehabilitation. This paper gives a quick overview of the state of the art, investigating video-based, wear-able, robotic, distributed, and gamified telerehabilitation solutions. In particular, agent-based solutions are analyzed and discussed addressing strength, limitations, and future challenges. Elaborating on functional requirements expressed by professional physiotherapists and researchers, the need for extending multi-agent systems (MAS) peculiarities at the sensing level in wearable solutions establishes new research challenges. Employed in cyber-physical scenarios with users-sensors and sensors-sensors interactions, MAS are requested to handle timing constraints, scarcity of resources and new communication means, which are crucial for providing real-time feedback and coaching.

Réalisations

Médias et communication
Nous contacter
Suivez la HES-SO
linkedin instagram facebook twitter youtube rss
univ-unita.eu www.eua.be swissuniversities.ch
Mentions légales
© 2021 - HES-SO.

HES-SO Rectorat