Davide Dell'Anna

@article{DBLP:journals/jair/DudzikWCDTBBSDEYWSKK24,
      title={Viewpoint: Hybrid Intelligence Supports Application Development for Diabetes Lifestyle Management}, 
      author={Bernd Dudzik and J. S. van der Waa and PY. Chen and R. Dobbe and I.M.R. de Troya and R. Bakker and M. H. T. de Boer and Q. Smit and Davide Dell'Anna and Emre Erdogan and Pinar Yolum and S. Wang and S. Baez Santamaria and L. Krause and B. A. Kamphorst},
      journal = {Journal of Artificial Intelligence Research (To Appear)},
      year={2024},
      url_Link = {/research.php},
      keywords  = {Hybrid Intelligence, Diabetes Lifestyle Management},
  abstract = {Type II diabetes is a complex health condition requiring patients to closely and continuously collaborate with healthcare professionals and other caretakers on lifestyle changes. While intelligent products have tremendous potential to support such Diabetes Lifestyle Management (DLM), existing products are typically conceived from a technology-centered perspective that insufficiently acknowledges the degree to which collaboration and inclusion of stakeholders is required. In this article, we argue that the emergent design philosophy of Hybrid Intelligence (HI) forms a suitable alternative lens for research and development. In particular, we (1) highlight a series of pragmatic challenges for effective AI-based DLM support based on results from an expert focus group, and (2) argue for HI’s potential to address these by outlining relevant research trajectories.}
}

Type II diabetes is a complex health condition requiring patients to closely and continuously collaborate with healthcare professionals and other caretakers on lifestyle changes. While intelligent products have tremendous potential to support such Diabetes Lifestyle Management (DLM), existing products are typically conceived from a technology-centered perspective that insufficiently acknowledges the degree to which collaboration and inclusion of stakeholders is required. In this article, we argue that the emergent design philosophy of Hybrid Intelligence (HI) forms a suitable alternative lens for research and development. In particular, we (1) highlight a series of pragmatic challenges for effective AI-based DLM support based on results from an expert focus group, and (2) argue for HI’s potential to address these by outlining relevant research trajectories.

@article{DBLP:journals/tosem/AbualhaijaADDFFF24,
      title={Replication in Requirements Engineering: the NLP for RE Case}, 
      author={Sallam Abualhaija and Fatma Başak Aydemir and Fabiano Dalpiaz and Davide Dell'Anna and Alessio Ferrari and Xavier Franch and Davide Fucci},
      journal      = {{ACM} Transactions on Software Engineering and Methodology},
      year={2024},
      url_Link = {https://dl.acm.org/doi/10.1145/3658669},
      url_Paper = {https://dl.acm.org/doi/pdf/10.1145/3658669},
      url_Supplement = {https://doi.org/10.6084/m9.figshare.21824481},
      keywords  = {Automated classification, Machine Learning, Software Engineering, Replication Study, Requirements Engineering, Software Testing, NLP4RE},
      abstract = {[Context] Natural language processing (NLP) techniques have been widely applied in the requirements engineering (RE) field to support tasks such as classification and ambiguity detection. Despite its empirical vocation, RE research has given limited attention to replication of NLP for RE studies. Replication is hampered by several factors, including the context specificity of the studies, the heterogeneity of the tasks involving NLP, the tasks’ inherent hairiness, and, in turn, the heterogeneous reporting structure. [Solution] To address these issues, we propose a new artifact, referred to as ID-Card, whose goal is to provide a structured summary of research papers emphasizing replication-relevant information. We construct the ID-Card through a structured, iterative process based on design science. [Results] In this paper: (i) we report on hands-on experiences of replication, (ii) we review the state-of-the-art and extract replication-relevant information, (iii) we identify, through focus groups, challenges across two typical dimensions of replication: data annotation and tool reconstruction, and (iv) we present the concept and structure of the ID-Card to mitigate the identified challenges. [Contribution] This study aims to create awareness of replication in NLP for RE. We propose an ID-Card that is intended to foster study replication, but can also be used in other contexts, e.g., for educational purposes.}
}

[Context] Natural language processing (NLP) techniques have been widely applied in the requirements engineering (RE) field to support tasks such as classification and ambiguity detection. Despite its empirical vocation, RE research has given limited attention to replication of NLP for RE studies. Replication is hampered by several factors, including the context specificity of the studies, the heterogeneity of the tasks involving NLP, the tasks’ inherent hairiness, and, in turn, the heterogeneous reporting structure. [Solution] To address these issues, we propose a new artifact, referred to as ID-Card, whose goal is to provide a structured summary of research papers emphasizing replication-relevant information. We construct the ID-Card through a structured, iterative process based on design science. [Results] In this paper: (i) we report on hands-on experiences of replication, (ii) we review the state-of-the-art and extract replication-relevant information, (iii) we identify, through focus groups, challenges across two typical dimensions of replication: data annotation and tool reconstruction, and (iv) we present the concept and structure of the ID-Card to mitigate the identified challenges. [Contribution] This study aims to create awareness of replication in NLP for RE. We propose an ID-Card that is intended to foster study replication, but can also be used in other contexts, e.g., for educational purposes.

@inproceedings{DBLP:conf/atal/DellAnnaMDGJOY24,
  author    = {Davide Dell'Anna and
               Pradeep K. Murukannaiah and
               Bernd Dudzik and
               Davide Grossi and
               Catholijn M. Jonker and
               Catharine Oertel and
               Pinar Yolum},
  title     = {Toward a Quality Model for Hybrid Intelligence Teams},
  booktitle = {Proceedings of the 23rd International Conference on Autonomous Agents
               and MultiAgent Systems, {AAMAS} 2024 (To Appear)},
  year = {2024},
  url_Link = {/research.php},
  url_Paper = {2024_AAMAS/AAMAS24_DellAnnaMDGJOY.pdf},
  url_Poster = {2024_AAMAS/AAMAS24_DellAnnaMDGJOY_Poster.pdf},
  url_Supplement = {https://doi.org/10.5281/zenodo.10593358},
  keywords  = {Hybrid Intelligence, Quality model, Human-agent teamwork, Sociotechnical systems, Team Diagnostic Survey},
  abstract = {Hybrid Intelligence (HI) is an emerging paradigm in which artificial intelligence (AI) augments human intelligence. The current literature lacks systematic models that guide the design and evaluation of HI systems. Further, discussions around HI primarily focus on technology, neglecting the holistic human-AI ensemble. In this paper, we take the initial steps toward the development of a quality model for characterizing and evaluating HI systems from a human-AI teams perspective. We conducted a study investigating the adequacy of properties commonly associated with effective human teams to describe HI. Our study, featuring the insights of 50 HI researchers, shows that various human team properties, including boundedness, interdependence, competency, purposefulness, initiative, normativity, and effectiveness, are important for HI systems. Our study also reveals limitations in applying certain human team properties, such as coaching, rewards, and recognition, to HI systems due to the inherent human-AI asymmetry.}
}

Hybrid Intelligence (HI) is an emerging paradigm in which artificial intelligence (AI) augments human intelligence. The current literature lacks systematic models that guide the design and evaluation of HI systems. Further, discussions around HI primarily focus on technology, neglecting the holistic human-AI ensemble. In this paper, we take the initial steps toward the development of a quality model for characterizing and evaluating HI systems from a human-AI teams perspective. We conducted a study investigating the adequacy of properties commonly associated with effective human teams to describe HI. Our study, featuring the insights of 50 HI researchers, shows that various human team properties, including boundedness, interdependence, competency, purposefulness, initiative, normativity, and effectiveness, are important for HI systems. Our study also reveals limitations in applying certain human team properties, such as coaching, rewards, and recognition, to HI systems due to the inherent human-AI asymmetry.

@article{DBLP:journals/ese/DellAnnaAD23,
author = {Davide Dell’Anna and
               Fatma Başak Aydemir and Fabiano Dalpiaz},
title = {Evaluating Classifiers in SE Research: The ECSER Pipeline and Two Replication Studies},
journal = {Empirical Software Engineering},
volume = {28},
number = {3},
year = {2023},
url_Link = {https://doi.org/10.1007/s10664-022-10243-1},
url_Paper = {https://link.springer.com/content/pdf/10.1007/s10664-022-10243-1.pdf},
url_Slides = {https://nlp4re.github.io/2023/assets/paper-templates/nlp4re-23-keynote-expanded.pdf},
url_Supplement = {https://doi.org/10.5281/zenodo.6266675},
doi       = {10.1007/s10664-022-10243-1},
keywords = {Automated classification, Machine Learning, Software Engineering, Replication Study, Requirements Engineering, Software Testing, NLP4RE},
abstract = {[Context] Automated classifiers, often based on machine learning (ML), are increasingly used in
software engineering (SE) for labelling previously unseen SE data. Researchers have proposed
automated classifiers that predict if a code chunk is a clone, if a requirement is functional or nonfunctional,
if the outcome of a test case is non-deterministic, etc. [Objective] The lack of guidelines for
applying and reporting classification techniques for SE research leads to studies in which important
research steps may be skipped, key findings might not be identified and shared, and the readers may
find reported results (e.g., precision or recall above 90%) that are not a credible representation of the
performance in operational contexts. The goal of this paper is to advance ML4SE research by proposing
rigorous ways of conducting and reporting research. We introduce the ECSER (Evaluating Classifiers in
Software Engineering Research) pipeline, which includes a series of steps for conducting and evaluating
automated classification research in SE. Then, we conduct two replication studies where we apply
ECSER to recent research in requirements engineering and in software testing. [Conclusions] In addition
to demonstrating the applicability of the pipeline, the replication studies demonstrate ECSER's
usefulness: not only do we confirm and strengthen some findings identified by the original authors, but
we also discover additional ones. Some of these findings contradict the original ones.}
}

[Context] Automated classifiers, often based on machine learning (ML), are increasingly used in software engineering (SE) for labelling previously unseen SE data. Researchers have proposed automated classifiers that predict if a code chunk is a clone, if a requirement is functional or nonfunctional, if the outcome of a test case is non-deterministic, etc. [Objective] The lack of guidelines for applying and reporting classification techniques for SE research leads to studies in which important research steps may be skipped, key findings might not be identified and shared, and the readers may find reported results (e.g., precision or recall above 90%) that are not a credible representation of the performance in operational contexts. The goal of this paper is to advance ML4SE research by proposing rigorous ways of conducting and reporting research. We introduce the ECSER (Evaluating Classifiers in Software Engineering Research) pipeline, which includes a series of steps for conducting and evaluating automated classification research in SE. Then, we conduct two replication studies where we apply ECSER to recent research in requirements engineering and in software testing. [Conclusions] In addition to demonstrating the applicability of the pipeline, the replication studies demonstrate ECSER's usefulness: not only do we confirm and strengthen some findings identified by the original authors, but we also discover additional ones. Some of these findings contradict the original ones.

@article{DBLP:journals/simulation/MooijBDDLS23,
author = {Jan de Mooij and
           Parantapa Bhattacharya and
           Davide Dell'Anna and
           Mehdi Dastani and
           Brian Logan and
           Samarth Swarup},
title = {A Framework for Modeling Human Behavior in Large-scale Agent-based Epidemic Simulations},
journal = {Simulation},
volume = {99},
number = {12},
pages = {1183--1211},
year = {2023},
url_Link = {https://doi.org/10.1177/00375497231184898},
url_Paper = {https://journals.sagepub.com/doi/epdf/10.1177/00375497231184898},
url_Video = {https://www.youtube.com/watch?v=MI63SK1FpKg},
url_Supplement = {https://github.com/A-Practical-Agent-Programming-Language/Sim-2APL},
doi = {10.1177/00375497231184898},
doi          = {10.1177/00375497231184898},
timestamp    = {Wed, 20 Dec 2023 09:58:01 +0100},
biburl       = {https://dblp.org/rec/journals/simulation/MooijBDDLS23.bib},
bibsource    = {dblp computer science bibliography, https://dblp.org},
keywords = {Agent-based modeling, Social simulation, Synthetic population, Computational epidemiology, COVID-19, PanSim, Sim-2APL, Large-Scale Agent-Based Simulation},
abstract = {Agent-based modeling is increasingly being used in computational epidemiology to characterize important behavioral
dimensions, such as the heterogeneity of the individual responses to interventions, when studying the spread of a
disease. Existing agent-based simulation frameworks and platforms currently fall in one of two categories: those that
can simulate millions of individuals with simple behaviors (e.g., based on simple state machines), and those that
consider more complex and social behaviors (e.g., agents that act according to their own agenda and preferences,
and deliberate about norm compliance) but, due to the computational complexity of reasoning involved, have limited
scalability. In this paper, we present a novel framework that enables large-scale distributed epidemic simulations with
complex behaving social agents whose decisions are based on a variety of concepts and internal attitudes such as
sense, knowledge, preferences, norms, and plans. The proposed framework supports simulations with millions of such
agents that can individually deliberate about their own knowledge, goals and preferences, and can adapt their behavior
based on other agents’ behaviors and on their attitude towards complying with norms. We showcase the applicability
and scalability of the proposed framework by developing a model of the spread of COVID-19 in the US state of Virginia.
Results illustrate that the framework can be effectively employed to simulate disease spreading with millions of complex
behaving agents and investigate behavioral interventions over a period of time of months.}
}

Agent-based modeling is increasingly being used in computational epidemiology to characterize important behavioral dimensions, such as the heterogeneity of the individual responses to interventions, when studying the spread of a disease. Existing agent-based simulation frameworks and platforms currently fall in one of two categories: those that can simulate millions of individuals with simple behaviors (e.g., based on simple state machines), and those that consider more complex and social behaviors (e.g., agents that act according to their own agenda and preferences, and deliberate about norm compliance) but, due to the computational complexity of reasoning involved, have limited scalability. In this paper, we present a novel framework that enables large-scale distributed epidemic simulations with complex behaving social agents whose decisions are based on a variety of concepts and internal attitudes such as sense, knowledge, preferences, norms, and plans. The proposed framework supports simulations with millions of such agents that can individually deliberate about their own knowledge, goals and preferences, and can adapt their behavior based on other agents’ behaviors and on their attitude towards complying with norms. We showcase the applicability and scalability of the proposed framework by developing a model of the spread of COVID-19 in the US state of Virginia. Results illustrate that the framework can be effectively employed to simulate disease spreading with millions of complex behaving agents and investigate behavioral interventions over a period of time of months.

@inproceedings{dellanna2023ddnr,
      title={Data-Driven Revision of Conditional Norms in Multi-Agent Systems (Extended Abstract)}, 
      author={Davide Dell'Anna and Natasha Alechina and Fabiano Dalpiaz and Mehdi Dastani and Brian Logan},
      booktitle = {Proceedings of the 32nd International Joint Conference on Artificial Intelligence, {IJCAI} 2023},
      year={2023},
      pages = {6868--6872},
      note = {Journal Track},
      url_Link = {https://doi.org/10.24963/ijcai.2023/773},
      url_Paper = {https://www.ijcai.org/proceedings/2023/0773.pdf},
      url_Slides      = {2023_IJCAI/IJCAI23_DellAnnaADDLL_Slides.pdf},
      url_Poster      = {2023_IJCAI/IJCAI23_DellAnnaADDLL_Poster.pdf},
      url_Video = {https://ijcai-23.org/video/?vid=39005627},
      url_Supplement = {https://doi.org/10.5281/zenodo.5907522},
      keywords  = {Data Driven Supervision Of Autonomous Systems, Norms, Multi-Agent Systems, Revision, Synthesis, Traffic Simulation},
  abstract = {In multi-agent systems, norm enforcement is a mechanism for steering the behavior of individual agents in order to achieve desired system-level objectives. Due to the dynamics of multi-agent systems, however, it is hard to design norms that guarantee the achievement of the objectives in every operating context. Also, these objectives may change over time, thereby making previously defined norms ineffective. In this paper, we investigate the use of system execution data to automatically synthesise and revise conditional prohibitions with deadlines, a type of norms aimed at preventing agents from exhibiting certain patterns of behaviors. We propose DDNR (Data-Driven Norm Revision), a data-driven approach to norm revision that synthesises revised norms with respect to a data set of traces describing the behavior of the agents in the system. We evaluate DDNR using a state-of-the-art, off-the-shelf urban traffic simulator. The results show that DDNR synthesises revised norms that are significantly more accurate than the original norms in distinguishing adequate and inadequate behaviors for the achievement of the system-level objectives.}
}

In multi-agent systems, norm enforcement is a mechanism for steering the behavior of individual agents in order to achieve desired system-level objectives. Due to the dynamics of multi-agent systems, however, it is hard to design norms that guarantee the achievement of the objectives in every operating context. Also, these objectives may change over time, thereby making previously defined norms ineffective. In this paper, we investigate the use of system execution data to automatically synthesise and revise conditional prohibitions with deadlines, a type of norms aimed at preventing agents from exhibiting certain patterns of behaviors. We propose DDNR (Data-Driven Norm Revision), a data-driven approach to norm revision that synthesises revised norms with respect to a data set of traces describing the behavior of the agents in the system. We evaluate DDNR using a state-of-the-art, off-the-shelf urban traffic simulator. The results show that DDNR synthesises revised norms that are significantly more accurate than the original norms in distinguishing adequate and inadequate behaviors for the achievement of the system-level objectives.

@techreport{abualhaija2023replication,
      title={Replication and Verifiability in Requirements Engineering: the NLP for RE Case}, 
      author={Sallam Abualhaija and Fatma Başak Aydemir and Fabiano Dalpiaz and Davide Dell'Anna and Alessio Ferrari and Xavier Franch and Davide Fucci},
      year={2023},
      eprint={2304.10265},
      archivePrefix={arXiv},
      primaryClass={cs.SE},
      url_Link       = {https://arxiv.org/abs/2304.10265},
      url_Paper = {https://arxiv.org/pdf/2304.10265.pdf},
      keywords  = {Automated classification, Machine Learning, Software Engineering, Replication Study, Requirements Engineering, Software Testing, NLP4RE},
  abstract = {[Context] Study replication is essential for theory building and empirical validation. [Problem] Despite its empirical vocation, requirements engineering (RE) research has given limited attention to study replication, threatening thereby the ability to verify existing results and use previous research as a baseline. [Solution] In this perspective paper, we -- a group of experts in natural language processing (NLP) for RE -- reflect on the challenges for study replication in NLP for RE. Concretely: (i) we report on hands-on experiences of replication, (ii) we review the state-of-the-art and extract replication-relevant information, and (iii) we identify, through focus groups, challenges across two typical dimensions of replication: data annotation and tool reconstruction. NLP for RE is a research area that is suitable for study replication since it builds on automated tools which can be shared, and quantitative evaluation that enable direct comparisons between results. [Results] Replication is hampered by several factors, including the context specificity of the studies, the heterogeneity of the tasks involving NLP, the tasks' inherent hairiness, and, in turn, the heterogeneous reporting structure. To address these issues, we propose an ID card whose goal is to provide a structured summary of research papers, with an emphasis on replication-relevant information. [Contribution] We contribute in this study with: (i) a set of reflections on replication in NLP for RE, (ii) a set of recommendations for researchers in the field to increase their awareness on the topic, and (iii) an ID card that is intended to primarily foster replication, and can also be used in other contexts, e.g., for educational purposes. Practitioners will also benefit from the results since replications increase confidence on research findings.}
}

[Context] Study replication is essential for theory building and empirical validation. [Problem] Despite its empirical vocation, requirements engineering (RE) research has given limited attention to study replication, threatening thereby the ability to verify existing results and use previous research as a baseline. [Solution] In this perspective paper, we – a group of experts in natural language processing (NLP) for RE – reflect on the challenges for study replication in NLP for RE. Concretely: (i) we report on hands-on experiences of replication, (ii) we review the state-of-the-art and extract replication-relevant information, and (iii) we identify, through focus groups, challenges across two typical dimensions of replication: data annotation and tool reconstruction. NLP for RE is a research area that is suitable for study replication since it builds on automated tools which can be shared, and quantitative evaluation that enable direct comparisons between results. [Results] Replication is hampered by several factors, including the context specificity of the studies, the heterogeneity of the tasks involving NLP, the tasks' inherent hairiness, and, in turn, the heterogeneous reporting structure. To address these issues, we propose an ID card whose goal is to provide a structured summary of research papers, with an emphasis on replication-relevant information. [Contribution] We contribute in this study with: (i) a set of reflections on replication in NLP for RE, (ii) a set of recommendations for researchers in the field to increase their awareness on the topic, and (iii) an ID card that is intended to primarily foster replication, and can also be used in other contexts, e.g., for educational purposes. Practitioners will also benefit from the results since replications increase confidence on research findings.

@article{DBLP:journals/eaai/DellAnnaJ22,
author = {Davide Dell’Anna and Anahita Jamshidnejad},
title = {Evolving Fuzzy Logic Systems for Creative Personalized Socially Assistive Robots},
journal = {Engineering Applications of Artificial Intelligence},
volume = {114},
pages = {105064},
year = {2022},
issn = {0952-1976},
url_Link = {https://doi.org/10.1016/j.engappai.2022.105064},
url_Paper = {https://www.sciencedirect.com/science/article/pii/S0952197622002251},
url_Supplement = {https://doi.org/10.5281/zenodo.6673720},
doi       = {10.1016/j.engappai.2022.105064},
  timestamp = {Wed, 10 Aug 2022 13:27:55 +0200},
  biburl    = {https://dblp.org/rec/journals/eaai/DellAnnaJ22.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
keywords = {Evolving Fuzzy logic Systems, Personalized Socially Assistive Robots, Robot Creativity, Socially Assistive Robots, Personalization, Creativity},
abstract = {Socially Assistive Robots (SARs) are increasingly used in dementia and elderly care. In order to provide effective assistance, SARs need to be personalized to individual patients and account for stimulating their divergent thinking in creative ways. Rule-based fuzzy logic systems provide effective methods for automated decision-making of SARs. However, expanding and modifying the rules of fuzzy logic systems to account for the evolving needs, preferences, and medical conditions of patients can be tedious and costly. In this paper, we introduce EFS4SAR, a novel Evolving Fuzzy logic System for Socially Assistive Robots that supports autonomous evolution of the fuzzy rules that steer the behavior of the SAR. EFS4SAR combines traditional rule-based fuzzy logic systems with evolutionary algorithms, which model the process of evolution in nature and have shown to result in creative behaviors. We evaluate EFS4SAR via computer simulations on both synthetic and real-world data. The results show that the fuzzy rules evolved over time are not only personalized with respect to the personal preferences and therapeutic needs of the patients, but they also meet the following criteria for creativity of SARs: originality and effectiveness of the therapeutic tasks proposed to the patients. Compared to existing evolving fuzzy systems, EFS4SAR achieves similar effectiveness with higher degree of originality.}
}

Socially Assistive Robots (SARs) are increasingly used in dementia and elderly care. In order to provide effective assistance, SARs need to be personalized to individual patients and account for stimulating their divergent thinking in creative ways. Rule-based fuzzy logic systems provide effective methods for automated decision-making of SARs. However, expanding and modifying the rules of fuzzy logic systems to account for the evolving needs, preferences, and medical conditions of patients can be tedious and costly. In this paper, we introduce EFS4SAR, a novel Evolving Fuzzy logic System for Socially Assistive Robots that supports autonomous evolution of the fuzzy rules that steer the behavior of the SAR. EFS4SAR combines traditional rule-based fuzzy logic systems with evolutionary algorithms, which model the process of evolution in nature and have shown to result in creative behaviors. We evaluate EFS4SAR via computer simulations on both synthetic and real-world data. The results show that the fuzzy rules evolved over time are not only personalized with respect to the personal preferences and therapeutic needs of the patients, but they also meet the following criteria for creativity of SARs: originality and effectiveness of the therapeutic tasks proposed to the patients. Compared to existing evolving fuzzy systems, EFS4SAR achieves similar effectiveness with higher degree of originality.

@article{DBLP:journals/jair/DellAnnaADDL22,
      title={Data-Driven Revision of Conditional Norms in Multi-Agent Systems}, 
      author={Davide Dell'Anna and Natasha Alechina and Fabiano Dalpiaz and Mehdi Dastani and Brian Logan},
      journal = {Journal of Artificial Intelligence Research},
      volume = {75},
      pages = {1549-1593},
      year={2022},
      url_Link = {https://doi.org/10.1613/jair.1.13683},
      url_Paper = {https://jair.org/index.php/jair/article/view/13683/26879},
      url_Slides      = {2023_IJCAI/IJCAI23_DellAnnaADDLL_Slides.pdf},
      url_Poster      = {2023_IJCAI/IJCAI23_DellAnnaADDLL_Poster.pdf},
        url_Video = {https://ijcai-23.org/video/?vid=39005627},
      url_Supplement = {https://doi.org/10.5281/zenodo.5907522},
      keywords  = {Data Driven Supervision Of Autonomous Systems, Norms, Multi-Agent Systems, Revision, Synthesis, Traffic Simulation},
  abstract = {In multi-agent systems, norm enforcement is a mechanism for steering the behavior of individual agents in order to achieve desired system-level objectives. Due to the dynamics
of multi-agent systems, however, it is hard to design norms that guarantee the achievement of the objectives in every operating context. Also, these objectives may change over time, thereby making previously defined norms ineffective. In this paper, we investigate the use of system execution data to automatically synthesise and revise conditional prohibitions with deadlines, a type of norms aimed at prohibiting agents from exhibiting certain patterns of behaviors. We propose DDNR (Data-Driven Norm Revision), a data-driven approach to norm revision that synthesises revised norms with respect to a data set of traces describing
the behavior of the agents in the system. We evaluate DDNR using a state-of-the-art, offthe-shelf urban traffic simulator. The results show that DDNR synthesises revised norms that are significantly more accurate than the original norms in distinguishing adequate and inadequate behaviors for the achievement of the system-level objectives.}
}

In multi-agent systems, norm enforcement is a mechanism for steering the behavior of individual agents in order to achieve desired system-level objectives. Due to the dynamics of multi-agent systems, however, it is hard to design norms that guarantee the achievement of the objectives in every operating context. Also, these objectives may change over time, thereby making previously defined norms ineffective. In this paper, we investigate the use of system execution data to automatically synthesise and revise conditional prohibitions with deadlines, a type of norms aimed at prohibiting agents from exhibiting certain patterns of behaviors. We propose DDNR (Data-Driven Norm Revision), a data-driven approach to norm revision that synthesises revised norms with respect to a data set of traces describing the behavior of the agents in the system. We evaluate DDNR using a state-of-the-art, offthe-shelf urban traffic simulator. The results show that DDNR synthesises revised norms that are significantly more accurate than the original norms in distinguishing adequate and inadequate behaviors for the achievement of the system-level objectives.

@inproceedings{dellanna2022complexity,
      title={The Complexity of Norm Synthesis and Revision}, 
      author={Davide Dell'Anna and Natasha Alechina and Fabiano Dalpiaz and Mehdi Dastani and Maarten Löffler and Brian Logan},
      booktitle = {Proceedings of the 15th International Workshop on Coordination, Organizations, Institutions, Norms, and Ethics for Governance
               of Multi-Agent Systems, {COINE@AAMAS} 2022},
      year={2022},
      url_Link = {https://doi.org/10.1007/978-3-031-20845-4_3},
      url_Paper      = {2022_COINE/COINE22_DellAnnaADDLL.pdf},
      url_Video = {https://www.youtube.com/watch?v=bU6tHw1KORc&t=9041s},
      keywords  = {Data Driven Supervision Of Autonomous Systems, Norms, Multi-Agent Systems, Revision, Synthesis, Complexity, NP Complete},
  abstract = {Norms have been widely proposed as a way of coordinating and controlling the activities of agents in a multi-agent system (MAS). A norm specifies the behaviour an agent should follow in order to achieve the objective of the MAS. However, designing norms to achieve a particular system objective can be difficult, particularly when there is no direct link between the language in which the system objective is stated and the language in which the norms can be expressed. In this paper, we consider the problem of synthesising a norm from traces of agent behaviour, where each trace is labelled with whether the behaviour satisfies the system objective. We show that the norm synthesis problem and several related problems are NP-complete.}
}

Norms have been widely proposed as a way of coordinating and controlling the activities of agents in a multi-agent system (MAS). A norm specifies the behaviour an agent should follow in order to achieve the objective of the MAS. However, designing norms to achieve a particular system objective can be difficult, particularly when there is no direct link between the language in which the system objective is stated and the language in which the norms can be expressed. In this paper, we consider the problem of synthesising a norm from traces of agent behaviour, where each trace is labelled with whether the behaviour satisfies the system objective. We show that the norm synthesis problem and several related problems are NP-complete.

@inproceedings{DBLP:conf/refsq/DalpiazDKM22,
  author    = {Fabiano Dalpiaz and
               Davide Dell'Anna and
               Sylwia Kopczynska and
               Lloyd Montgomery},
  title     = {Preface: 5th Workshop on Natural Language Processing for Requirements
               Engineering {(NLP4RE)}},
  booktitle = {Joint Proceedings of {REFSQ-2022} Workshops, Doctoral Symposium, and
               Posters {\&} Tools Track co-located with the 28th International
               Conference on Requirements Engineering: Foundation for Software Quality,
               {REFSQ} 2022},
  volume    = {3122},
  publisher = {CEUR-WS.org},
  year      = {2022},
  url_Link       = {http://ceur-ws.org/Vol-3122/NLP4RE-preface.pdf},
  timestamp = {Thu, 21 Apr 2022 17:13:26 +0200},
  biburl    = {https://dblp.org/rec/conf/refsq/DalpiazDKM22.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@proceedings{DBLP:conf/refsq/2022w,
  editor    = {Jannik Fischbach and
               Nelly Condori{-}Fern{\'{a}}ndez and
               J{\"{o}}rg D{\"{o}}rr and
               Marcela Ruiz and
               Jan{-}Philipp Stegh{\"{o}}fer and
               Liliana Pasquale and
               Andrea Zisman and
               Renata S. S. Guizzardi and
               Jennifer Horkoff and
               Anna Perini and
               Angelo Susi and
               Maya Daneva and
               Andrea Herrmann and
               Kurt Schneider and
               Patrick Mennig and
               Fabiano Dalpiaz and
               Davide Dell'Anna and
               Sylwia Kopczynska and
               Lloyd Montgomery and
               Andy G. Darby and
               Peter Sawyer},
  title     = {Joint Proceedings of {REFSQ-2022} Workshops, Doctoral Symposium, and
               Posters {\&} Tools Track co-located with the 28th International
               Conference on Requirements Engineering: Foundation for Software Quality
               {(REFSQ} 2022), Aston, Birmingham, UK, March 21, 2022},
  series    = {{CEUR} Workshop Proceedings},
  volume    = {3122},
  publisher = {CEUR-WS.org},
  year      = {2022},
  url_Link       = {http://ceur-ws.org/Vol-3122},
  urn       = {urn:nbn:de:0074-3122-3},
  timestamp = {Thu, 21 Apr 2022 17:13:26 +0200},
  biburl    = {https://dblp.org/rec/conf/refsq/2022w.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@phdthesis{DBLP:phd/basesearch/DellAnna21,
  author    = {Davide Dell'Anna},
  title     = {Data-Driven Supervision of Autonomous Systems},
  school    = {Utrecht University, Netherlands},
  year      = {2021},
  url_Link       = {https://doi.org/10.33540/589},
  url_Paper       = {2021_PHDTHESIS/DellAnna_2021_PhDThesis.pdf},
  doi       = {10.33540/589},
  isbn      = {978-94-6416-395-7},
  keywords  = {Data Driven Supervision Of Autonomous Systems},
  abstract = {Modern software systems execute in increasingly dynamic settings, and their objectives are in constant motion. In order to preserve their adequacy and effectiveness within an evolving environment, software and its requirements need to adapt to change. In this dissertation, we propose a data-driven supervision framework for the automatic run-time revision of requirements, so to ensure the achievement of system-level objectives in dynamic settings. We focus on the supervision of multi-agent systems (MASs), collections of interacting autonomous agents, such as autonomous cars on smart roads. In multi-agent systems, agents’ internals are typically unknown to the other agents and to the MAS designer. Norms are often employed as a means for controlling and coordinating the agents' behavior without over-constraining their autonomy. We use norms to characterize requirements for the behavior of the agents in the system, and we use sanctions as a deterrence mechanism to discourage agents from violations. The proposed supervision framework employs a general architecture for system self-adaptation, described as a closed control-loop. At run-time, the system is monitored and execution data is collected in different operating contexts. The collected data is used to learn statistical correlations between the achievement of the system's objectives and the satisfaction of the requirements in the different operating contexts. The learnt information is applied to automatically assess the validity of the assumptions made at design-time, and to automatically synthesise new requirements and sanctions when there is evidence that the current ones are not effective.}
}

Modern software systems execute in increasingly dynamic settings, and their objectives are in constant motion. In order to preserve their adequacy and effectiveness within an evolving environment, software and its requirements need to adapt to change. In this dissertation, we propose a data-driven supervision framework for the automatic run-time revision of requirements, so to ensure the achievement of system-level objectives in dynamic settings. We focus on the supervision of multi-agent systems (MASs), collections of interacting autonomous agents, such as autonomous cars on smart roads. In multi-agent systems, agents’ internals are typically unknown to the other agents and to the MAS designer. Norms are often employed as a means for controlling and coordinating the agents' behavior without over-constraining their autonomy. We use norms to characterize requirements for the behavior of the agents in the system, and we use sanctions as a deterrence mechanism to discourage agents from violations. The proposed supervision framework employs a general architecture for system self-adaptation, described as a closed control-loop. At run-time, the system is monitored and execution data is collected in different operating contexts. The collected data is used to learn statistical correlations between the achievement of the system's objectives and the satisfaction of the requirements in the different operating contexts. The learnt information is applied to automatically assess the validity of the assumptions made at design-time, and to automatically synthesise new requirements and sanctions when there is evidence that the current ones are not effective.

@inproceedings{bhattacharya21pansim,
  author    = {Parantapa Bhattacharya and Jan {de Mooij} and Davide Dell'Anna and Mehdi Dastani and Brian Logan and Samarth Swarup},
  title     = {Pan{S}im + {Sim-2APL}: {A} Platform for Large-Scale Distributed Simulation with Complex Agents},
  booktitle = {Proceedings of the 9th International Workshop on Engineering Multi-Agent Systems, {EMAS@AAMAS} 2021},
  volume    = {13190},
  pages     = {1--21},
  year      = {2021},
  url_Link = {https://doi.org/10.1007/978-3-030-97457-2_1},
  url_Paper      = {2021_EMAS/EMAS21_BhattacharyaMDDLS.pdf},
  url_Slides = {2021_EMAS/EMAS21_BhattacharyaMDDLS_Slides.pdf},
  url_Video = {https://vimeo.com/539341278},
  url_Supplement       = {https://bitbucket.org/goldenagents/sim2apl-episimpledemics/src/master/},
  doi = {10.1007/978-3-030-97457-2_1},
  keywords  = {Distributed simulation, Agent-based simulation, Social Simulation, Large-Scale Agent-Based Simulation},
  abstract = {Agent-based simulation is increasingly being used to model social phenomena involving large numbers of agents. However,
existing agent-based simulation platforms severely limit the kinds of the social phenomena that can modeled, as they do not
support large scale simulations involving agents with complex behaviors. In this paper, we present a scalable agent-based
simulation framework that supports modelling of complex social phenomena. The framework integrates a new simulation platform
that exploits distributed computer architectures, with an extension of a multi-agent programming technology that allows
development of complex deliberative agents. To show the scalability of our framework, we briefly describe its application to
the development of a model of the spread of COVID-19 involving complex deliberative agents in the US state of Virginia.}
}

Agent-based simulation is increasingly being used to model social phenomena involving large numbers of agents. However, existing agent-based simulation platforms severely limit the kinds of the social phenomena that can modeled, as they do not support large scale simulations involving agents with complex behaviors. In this paper, we present a scalable agent-based simulation framework that supports modelling of complex social phenomena. The framework integrates a new simulation platform that exploits distributed computer architectures, with an extension of a multi-agent programming technology that allows development of complex deliberative agents. To show the scalability of our framework, we briefly describe its application to the development of a model of the spread of COVID-19 involving complex deliberative agents in the US state of Virginia.

@inproceedings{DBLP:conf/ampm/MooijDBDLS21,
  author    = {Jan de Mooij and
               Davide Dell'Anna and
               Parantapa Bhattacharya and
               Mehdi Dastani and
               Brian Logan and
               Samarth Swarup},
  title     = {Using Agent-Based Simulation to Investigate Behavioral Interventions in a Pandemic},
  booktitle = {Proceedings of the 1st Workshop on Agent-based Modelling and Policy-Making, {AMPM@JURIX} 2021},
  year      = {2021},
  url_Paper = {http://ceur-ws.org/Vol-3182/paper7.pdf},
   url_Slides       = {https://github.com/ampmresearch/ampmresearch.github.io/blob/main/presentations/7_JdeMooij.pdf},
  keywords  = {COVID-19, synthetic population, Sim-2APL, BDI, multiagent systems, agent-based simulation, norms, executive orders, behavior, Large-Scale Agent-Based Simulation},
  abstract = {Simulation is a useful tool for evaluating behavioral interventions when adoption rate among the population may be uncertain. Individual agent models are often prohibitively expensive, but unlike stochastic models allow studying compliance heterogeneity.
In this paper we aim to demonstrate the feasibility of evaluating behavioral intervention policies using large-scale data-driven agent-based simulations. We explain how the simulation is calibrated with respect to real-world data, and demonstrate its utility by studying the effectiveness of interventions used in Virginia in early 2020 through counterfactual simulations.}
}

Simulation is a useful tool for evaluating behavioral interventions when adoption rate among the population may be uncertain. Individual agent models are often prohibitively expensive, but unlike stochastic models allow studying compliance heterogeneity. In this paper we aim to demonstrate the feasibility of evaluating behavioral intervention policies using large-scale data-driven agent-based simulations. We explain how the simulation is calibrated with respect to real-world data, and demonstrate its utility by studying the effectiveness of interventions used in Virginia in early 2020 through counterfactual simulations.

@inproceedings{DBLP:conf/mabs/MooijDBDLS21,
  author    = {Jan de Mooij and
               Davide Dell'Anna and
               Parantapa Bhattacharya and
               Mehdi Dastani and
               Brian Logan and
               Samarth Swarup},
  title     = {Quantifying the Effects of Norms on {COVID-19} Cases Using an Agent-Based
               Simulation},
  booktitle = {Proceedings of the 22nd International Workshop on Multi-Agent Systems and Agent-Based Simulation, {MABS@AAMAS} 2021},
  volume    = {13128},
  pages     = {99--112},
  year      = {2021},
  url_Link = {https://doi.org/10.1007/978-3-030-94548-0_8},
  url_Paper      = {2021_MABS/MABS21_deMooijDBDLS.pdf},
  url_Supplement       = {https://bitbucket.org/goldenagents/sim2apl-episimpledemics/src/master/},
  doi       = {10.1007/978-3-030-94548-0\_8},
  timestamp = {Wed, 19 Jan 2022 09:36:09 +0100},
  biburl    = {https://dblp.org/rec/conf/mabs/MooijDBDLS21.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  keywords  = {COVID-19, synthetic population, Sim-2APL, BDI, multiagent systems, agent-based simulation, norms, executive orders, behavior, Large-Scale Agent-Based Simulation},
  abstract = {Modelling social phenomena in large-scale agent-based simulations has long been a challenge due to the computational cost of incorporating agents whose behaviors are determined by reasoning about their internal attitudes and external factors. However, COVID-19 has brought the urgency of doing this to the fore, as, in the absence of viable pharmaceutical interventions, the progression of the pandemic has primarily been driven by behaviors and behavioral interventions. In this paper, we address this problem by developing a large-scale data-driven agent-based simulation model where individual agents reason about their beliefs, objectives, trust in government, and the norms imposed by the government. These internal and external attitudes are based on actual data concerning daily activities of individuals, their political orientation, and norms being enforced in the US state of Virginia. Our model is calibrated and validated using mobility and COVID-19 case data. We show the utility of our model by quantifying the benefits of the various behavioral interventions through counterfactual runs of our calibrated simulation.}
}

Modelling social phenomena in large-scale agent-based simulations has long been a challenge due to the computational cost of incorporating agents whose behaviors are determined by reasoning about their internal attitudes and external factors. However, COVID-19 has brought the urgency of doing this to the fore, as, in the absence of viable pharmaceutical interventions, the progression of the pandemic has primarily been driven by behaviors and behavioral interventions. In this paper, we address this problem by developing a large-scale data-driven agent-based simulation model where individual agents reason about their beliefs, objectives, trust in government, and the norms imposed by the government. These internal and external attitudes are based on actual data concerning daily activities of individuals, their political orientation, and norms being enforced in the US state of Virginia. Our model is calibrated and validated using mobility and COVID-19 case data. We show the utility of our model by quantifying the benefits of the various behavioral interventions through counterfactual runs of our calibrated simulation.

@techreport{dellanna2021complexity,
      title={The Complexity of Data-Driven Norm Synthesis and Revision}, 
      author={Davide Dell'Anna and Natasha Alechina and Brian Logan and Maarten Löffler and Fabiano Dalpiaz and Mehdi Dastani},
      year={2021},
      eprint={2112.02626},
      archivePrefix={arXiv},
      primaryClass={cs.CC},
      url_Paper       = {https://arxiv.org/abs/2112.02626},
      keywords  = {Data Driven Supervision Of Autonomous Systems, Norms, Multi-Agent Systems, Revision, Synthesis, Complexity, NP Complete},
  abstract = {Norms have been widely proposed as a way of coordinating and controlling the activities of agents in a multi-agent system (MAS). A norm specifies the behaviour an agent should follow in order to achieve the objective of the MAS. However, designing norms to achieve a particular system objective can be difficult, particularly when there is no direct link between the language in which the system objective is stated and the language in which the norms can be expressed. In this paper, we consider the problem of synthesising a norm from traces of agent behaviour, where each trace is labelled with whether the behaviour satisfies the system objective. We show that the norm synthesis problem is NP-complete.}
}

Norms have been widely proposed as a way of coordinating and controlling the activities of agents in a multi-agent system (MAS). A norm specifies the behaviour an agent should follow in order to achieve the objective of the MAS. However, designing norms to achieve a particular system objective can be difficult, particularly when there is no direct link between the language in which the system objective is stated and the language in which the norms can be expressed. In this paper, we consider the problem of synthesising a norm from traces of agent behaviour, where each trace is labelled with whether the behaviour satisfies the system objective. We show that the norm synthesis problem is NP-complete.

@article{DBLP:journals/aamas/DellAnnaDD20,
  author    = {Davide Dell'Anna and
               Mehdi Dastani and
               Fabiano Dalpiaz},
  title     = {Runtime revision of sanctions in normative multi-agent systems},
  journal   = {Autonomous Agents and Multi-Agent Systems.},
  volume    = {34},
  number    = {2},
  pages     = {43},
  year      = {2020},
  url_Paper       = {https://doi.org/10.1007/s10458-020-09465-8},
  url_Supplement       = {https://zenodo.org/record/3712045#.XwihvSgzaUk},
  doi       = {10.1007/s10458-020-09465-8},
  keywords  = {Norm Revision, Norms, MAS, Multi-Agent Systems, Sanctions, Preferences, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Thu, 06 Aug 2020 01:00:00 +0200},
  biburl    = {https://dblp.org/rec/journals/aamas/DellAnnaDD20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {To achieve system-level properties of a multiagent system, the behavior of individual agents should be controlled and coordinated. One way to control agents without limiting their autonomy is to enforce norms by means of sanctions. The dynamicity and unpredictability of the agents' interactions in uncertain environments, however, make it hard for designers to specify norms that will guarantee the achievement of the systemlevel objectives in every operating context. 
                  In this paper, we propose a runtime mechanism for the automated revision of norms by altering their sanctions. We use a Bayesian Network to learn, from system execution data, the relationship between the obedience/violation of the norms and the achievement of the system-level objectives. By combining the knowledge acquired at runtime with an estimation of the preferences of rational agents, we devise heuristic strategies that automatically revise the sanctions of the enforced norms. We evaluate our heuristics using a traffic simulator and we show that our mechanism is able to quickly identify optimal revisions of the initially enforced norms.}
}

To achieve system-level properties of a multiagent system, the behavior of individual agents should be controlled and coordinated. One way to control agents without limiting their autonomy is to enforce norms by means of sanctions. The dynamicity and unpredictability of the agents' interactions in uncertain environments, however, make it hard for designers to specify norms that will guarantee the achievement of the systemlevel objectives in every operating context. In this paper, we propose a runtime mechanism for the automated revision of norms by altering their sanctions. We use a Bayesian Network to learn, from system execution data, the relationship between the obedience/violation of the norms and the achievement of the system-level objectives. By combining the knowledge acquired at runtime with an estimation of the preferences of rational agents, we devise heuristic strategies that automatically revise the sanctions of the enforced norms. We evaluate our heuristics using a traffic simulator and we show that our mechanism is able to quickly identify optimal revisions of the initially enforced norms.

@article{DBLP:journals/ase/DellAnnaDD19,
  author    = {Davide Dell'Anna and
               Fabiano Dalpiaz and
               Mehdi Dastani},
  title     = {Requirements-driven evolution of sociotechnical systems via probabilistic
               reasoning and hill climbing},
  journal   = {Automated Software Engineering},
  volume    = {26},
  number    = {3},
  pages     = {513--557},
  year      = {2019},
  url_Paper       = {https://doi.org/10.1007/s10515-019-00255-5},
  doi       = {10.1007/s10515-019-00255-5},
  keywords  = {Requirements, Revision, Evolution, Socio-Technical Systems, Bayesian Networks, Hill-Climbing, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Thu, 31 Oct 2019 00:00:00 +0100},
  biburl    = {https://dblp.org/rec/journals/ase/DellAnnaDD19.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {Sociotechnical systems (STSs) are defined by the interaction between technical systems, like software and machines, and social entities, like humans and organizations. 
                  The entities within an STS are autonomous, thus weakly controllable, and the environment where the STS operates is highly dynamic.
                  As a result, the design artifacts that represent the requirements of an STS, such as requirements models, may end up being invalid when the system operates, for the autonomous entities do not comply with the requirements, or the environment changes.
                  In this paper, we present a framework that uses runtime execution data to support the runtime validation of requirements models and to guide the evolution of an STS.
                  We propose two types of evolution: (i) <i>manual</i>: the analyst uses Bayesian inference to discover which assumptions in a requirements model are invalid and manually adjusts the system or its model; and (ii) <i>automated</i>: requirements are iteratively revised by an hill climbing algorithm searching for requirements that maximize the achievement of the stakeholders' objectives. 
                  We evaluate the effectiveness of different revision heuristics on a smart traffic simulation applied to an exemplar from the self-adaptive systems literature.
                  The results show that our heuristics, informed by runtime execution data, outperform standard uninformed heuristics, in terms of convergence speed, solution quality, and stability. Moreover, the algorithms show good resilience to noise introduced into the execution data.}
}

Sociotechnical systems (STSs) are defined by the interaction between technical systems, like software and machines, and social entities, like humans and organizations. The entities within an STS are autonomous, thus weakly controllable, and the environment where the STS operates is highly dynamic. As a result, the design artifacts that represent the requirements of an STS, such as requirements models, may end up being invalid when the system operates, for the autonomous entities do not comply with the requirements, or the environment changes. In this paper, we present a framework that uses runtime execution data to support the runtime validation of requirements models and to guide the evolution of an STS. We propose two types of evolution: (i) manual: the analyst uses Bayesian inference to discover which assumptions in a requirements model are invalid and manually adjusts the system or its model; and (ii) automated: requirements are iteratively revised by an hill climbing algorithm searching for requirements that maximize the achievement of the stakeholders' objectives. We evaluate the effectiveness of different revision heuristics on a smart traffic simulation applied to an exemplar from the self-adaptive systems literature. The results show that our heuristics, informed by runtime execution data, outperform standard uninformed heuristics, in terms of convergence speed, solution quality, and stability. Moreover, the algorithms show good resilience to noise introduced into the execution data.

@inproceedings{DBLP:conf/bnaic/DellAnnaDD19,
  author       = {Davide Dell'Anna and
                  Mehdi Dastani and
                  Fabiano Dalpiaz},
  title        = {Runtime Revision of Norms and Sanctions Based on Agent Preferences (Extended Abstract)},
  booktitle    = {Proceedings of the 31st Benelux Conference on Artificial Intelligence
                  {(BNAIC} 2019) and the 28th Belgian Dutch Conference on Machine Learning
                  (Benelearn 2019)},
  series       = {{CEUR} Workshop Proceedings},
  volume       = {2491},
  publisher    = {CEUR-WS.org},
  year         = {2019},
  url          = {https://ceur-ws.org/Vol-2491/abstract43.pdf},
  timestamp    = {Fri, 10 Mar 2023 16:22:44 +0100},
  biburl       = {https://dblp.org/rec/conf/bnaic/DellAnnaDD19.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

@inproceedings{DBLP:conf/atal/DellAnnaDD19,
  author    = {Davide Dell'Anna and
               Mehdi Dastani and
               Fabiano Dalpiaz},
  title     = {Runtime Revision of Norms and Sanctions based on Agent Preferences},
  booktitle = {Proceedings of the 18th International Conference on Autonomous Agents
               and MultiAgent Systems, {AAMAS} 2019},
  pages     = {1609--1617},
  year      = {2019},
  url_Link       = {http://dl.acm.org/citation.cfm?id=3331881},
  url_Paper      = {2019_AAMAS/AAMAS19_DellAnna.pdf},
  url_Slides= {2019_AAMAS/AAMAS19_DellAnna_Slides.pdf},
  url_Poster= {2019_AAMAS/AAMAS19_DellAnna_Poster.pdf},
  keywords  = {Norm Revision, Norms, MAS, Multi-Agent Systems, Sanctions, Preferences, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Wed, 29 May 2019 16:36:58 +0200},
  biburl    = {https://dblp.org/rec/conf/atal/DellAnnaDD19.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {To fulfill the overall objectives of a multiagent system, the behavior of individual agents should be controlled and coordinated. Runtime norm enforcement is one way to do so without over-constraining the agents' autonomy. Due to the dynamicity and uncertainty of the environment, however, it is hard to specify norms that, when enforced, will fulfill the system-level objectives in every operating context. In this paper, we propose a mechanism for the automated revision of norms by altering their sanctions, based on the data monitored during the system execution and on some knowledge about the agents' preferences. We use a Bayesian Network to learn at runtime the relationship between the obedience of a norm and the achievement of the system objectives. We propose two heuristic strategies that explore the updated Bayesian Network and automatically revise the sanction of an enforced norm. An evaluation of our heuristics using a traffic simulator shows that our mechanisms outperform uninformed heuristics in terms of convergence speed.}
}

To fulfill the overall objectives of a multiagent system, the behavior of individual agents should be controlled and coordinated. Runtime norm enforcement is one way to do so without over-constraining the agents' autonomy. Due to the dynamicity and uncertainty of the environment, however, it is hard to specify norms that, when enforced, will fulfill the system-level objectives in every operating context. In this paper, we propose a mechanism for the automated revision of norms by altering their sanctions, based on the data monitored during the system execution and on some knowledge about the agents' preferences. We use a Bayesian Network to learn at runtime the relationship between the obedience of a norm and the achievement of the system objectives. We propose two heuristic strategies that explore the updated Bayesian Network and automatically revise the sanction of an enforced norm. An evaluation of our heuristics using a traffic simulator shows that our mechanisms outperform uninformed heuristics in terms of convergence speed.

@inproceedings{DBLP:conf/re/DalpiazDAC19,
  author    = {Fabiano Dalpiaz and
               Davide Dell'Anna and
               Fatma Başak Aydemir and
               Sercan {\c{C}}evikol},
  title     = {Requirements Classification with Interpretable Machine Learning and
               Dependency Parsing},
  booktitle = {Proceedings of the 27th {IEEE} International Requirements Engineering Conference, {RE}
               2019},
  pages     = {142--152},
  year      = {2019},
  url_Link       = {https://doi.org/10.1109/RE.2019.00025},
  url_Paper      = {2019_RE/RE19_Dalpiaz.pdf},
  url_Slides= {2019_RE/RE19_Dalpiaz_Slides.pdf},
  url_Supplement = {https://doi.org/10.5281/zenodo.3309582},
  doi       = {10.1109/RE.2019.00025},
  keywords  = {Requirements, Requirements Classification, Machine Learning, Classification, Dependency Parsing, NLP4RE},
  timestamp = {Mon, 15 Jun 2020 01:00:00 +0200},
  biburl    = {https://dblp.org/rec/conf/re/DalpiazDAC19.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {Requirements classification is a traditional application of machine learning (ML) to RE that helps handle large requirements datasets. A prime example of an RE classification problem is the distinction between functional and non-functional (quality) requirements. 
                  State-of-the-art classifiers build their effectiveness on a large set of word features like text n-grams or POS n-grams, which do not fully capture the essence of a requirement. As a result, it is arduous for human analysts to interpret the classification results by exploring the classifier's inner workings. 
                  We propose the use of more general linguistic features, such as dependency types, for the construction of interpretable ML classifiers for RE. Through a feature engineering effort, in which we are assisted by modern introspection tools that reveal the hidden inner workings of ML classifiers, we derive a set of 17 linguistic features.
                  While classifiers that use our proposed features fit the training set slightly worse than those that use high-dimensional feature sets, our approach performs generally better on validation datasets and it is more interpretable.}
}

Requirements classification is a traditional application of machine learning (ML) to RE that helps handle large requirements datasets. A prime example of an RE classification problem is the distinction between functional and non-functional (quality) requirements. State-of-the-art classifiers build their effectiveness on a large set of word features like text n-grams or POS n-grams, which do not fully capture the essence of a requirement. As a result, it is arduous for human analysts to interpret the classification results by exploring the classifier's inner workings. We propose the use of more general linguistic features, such as dependency types, for the construction of interpretable ML classifiers for RE. Through a feature engineering effort, in which we are assisted by modern introspection tools that reveal the hidden inner workings of ML classifiers, we derive a set of 17 linguistic features. While classifiers that use our proposed features fit the training set slightly worse than those that use high-dimensional feature sets, our approach performs generally better on validation datasets and it is more interpretable.

@inproceedings{DBLP:conf/prima/DellAnnaDD18,
  author    = {Davide Dell'Anna and
               Mehdi Dastani and
               Fabiano Dalpiaz},
  title     = {Runtime Norm Revision Using Bayesian Networks},
  booktitle = {Proceedings of the 21st International Conference on Principles and Practice of Multi-Agent Systems, {PRIMA} 2018},
  volume    = {11224},
  pages     = {279--295},
  year      = {2018},
  url_Link       = {https://doi.org/10.1007/978-3-030-03098-8\_17},
  url_Paper      = {2018_PRIMA/PRIMA18_DellAnna.pdf},
  url_Slides= {2018_PRIMA/PRIMA18_DellAnna_Slides.pdf},
  url_Poster= {2018_PRIMA/PRIMA18_DellAnna_Poster.pdf},
  doi       = {10.1007/978-3-030-03098-8\_17},
  keywords  = {Norm Revision, Norms, Bayesian Networks, Runtime, MAS, Multi-Agent Systems, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Mon, 16 Sep 2019 01:00:00 +0200},
  biburl    = {https://dblp.org/rec/conf/prima/DellAnnaDD18.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {To guarantee the overall desirable objectives of multiagent systems, the behavior of individual agents should be controlled and coordinated. Runtime norm enforcement is a mechanism to control and coordinate the behavior of agents at runtime without limiting their autonomy. However, due to the dynamicity and uncertainty involved in the agents' environments, the enforced norms can sometimes be ineffective. In this paper we propose a runtime supervision mechanism that automatically revises norms when their enforcement appears to be ineffective. The decision to revise norms is taken based on a Bayesian Network that gives information about the likelihood of achieving the overall desirable system objectives by enforcing the norms. Norms can be revised in three ways: relaxation, strengthening and alteration. We evaluate the supervision mechanism based on an urban smart traffic simulation.}
}

To guarantee the overall desirable objectives of multiagent systems, the behavior of individual agents should be controlled and coordinated. Runtime norm enforcement is a mechanism to control and coordinate the behavior of agents at runtime without limiting their autonomy. However, due to the dynamicity and uncertainty involved in the agents' environments, the enforced norms can sometimes be ineffective. In this paper we propose a runtime supervision mechanism that automatically revises norms when their enforcement appears to be ineffective. The decision to revise norms is taken based on a Bayesian Network that gives information about the likelihood of achieving the overall desirable system objectives by enforcing the norms. Norms can be revised in three ways: relaxation, strengthening and alteration. We evaluate the supervision mechanism based on an urban smart traffic simulation.

@inproceedings{DBLP:conf/re/DellAnnaDD18,
  author    = {Davide Dell'Anna and
               Fabiano Dalpiaz and
               Mehdi Dastani},
  title     = {Validating Goal Models via Bayesian Networks},
  booktitle = {Proceedings of the 5th International Workshop on Artificial Intelligence for Requirements
               Engineering, AIRE@RE 2018},
  pages     = {39--46},
  year      = {2018},
  url_Link       = {https://doi.org/10.1109/AIRE.2018.00012},
  url_Paper      = {2018_AIRE/AIRE18_DellAnna.pdf},
  url_Slides= {2018_AIRE/AIRE18_DellAnna_Slides.pdf},
  doi       = {10.1109/AIRE.2018.00012},
  keywords  = {Goal Models, Bayesian Networks, Assumptions, Validation, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Mon, 15 Jun 2020 01:00:00 +0200},
  biburl    = {https://dblp.org/rec/conf/re/DellAnnaDD18.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {Goal models are an example of requirement modeling language that has been applied to support the runtime monitoring and diagnosis of software systems and to steer self-adaptive systems. When creating a goal model, requirement engineers make <i>assumptions</i> concerning how the goals relate to each other and when they should be considered as satisfied. In dynamic environments, however, the assumptions made in the model may be (or become) invalid. This may result in a system that does not satisfy the stakeholders' needs and, when the model is used in adaptive systems, ineffective reconfigurations. Only few and preliminary works address the automated validation of goal or requirement models. In this paper we propose the use of probabilistic models (Bayesian Networks) to determine the validity of the assumptions underlying a goal model. We employ empirical data and probabilistic inference to automatically determine a quantitative degree of validity of goal model assumptions. We illustrate the approach on a smart traffic scenario.}
}

Goal models are an example of requirement modeling language that has been applied to support the runtime monitoring and diagnosis of software systems and to steer self-adaptive systems. When creating a goal model, requirement engineers make assumptions concerning how the goals relate to each other and when they should be considered as satisfied. In dynamic environments, however, the assumptions made in the model may be (or become) invalid. This may result in a system that does not satisfy the stakeholders' needs and, when the model is used in adaptive systems, ineffective reconfigurations. Only few and preliminary works address the automated validation of goal or requirement models. In this paper we propose the use of probabilistic models (Bayesian Networks) to determine the validity of the assumptions underlying a goal model. We employ empirical data and probabilistic inference to automatically determine a quantitative degree of validity of goal model assumptions. We illustrate the approach on a smart traffic scenario.

@inproceedings{DBLP:conf/refsq/DellAnna18,
  author    = {Davide Dell'Anna},
  title     = {Requirements-Driven Supervision of Socio-Technical Systems},
  booktitle = {Joint Proceedings of {REFSQ-2018} Workshops, Doctoral Symposium, Live
               Studies Track, and Poster Track, {REFSQ} 2018},
  volume    = {2075},
  year      = {2018},
  url       = {http://ceur-ws.org/Vol-2075/DS-paper1.pdf},
  url_Slides= {2018_REFSQ/REFSQ18_DellAnna_Slides.pdf},
  keywords  = {NRequirements, Revision, Supervision, Socio-Technical Systems, Data Driven Supervision Of Autonomous Systems},
  timestamp = {Fri, 17 Jul 2020 11:59:20 +0200},
  biburl    = {https://dblp.org/rec/conf/refsq/DellAnna18.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {Modern software systems are characterized by ever-changing goals and requirements. Such systems operate in an environment that is dynamic, open, partly known, unpredictable. New goals arise and others are dropped, due to changes in stakeholders' needs and priorities, government regulations, technology. Despite this dynamism, systems should meet their goals and comply with the evolving requirements.
                  While several self-adaptation mechanisms have been proposed in the literature, they cannot be fully applied for socio-technical systems that involve autonomous (thus, non-controllable) components.
                  This project aims at designing and developing a runtime requirements supervision framework that monitors the execution of socio-technical systems, evaluates their behavior against the overall goals and intervenes by deciding how to revise requirements when adaptation is not possible.}
}

Modern software systems are characterized by ever-changing goals and requirements. Such systems operate in an environment that is dynamic, open, partly known, unpredictable. New goals arise and others are dropped, due to changes in stakeholders' needs and priorities, government regulations, technology. Despite this dynamism, systems should meet their goals and comply with the evolving requirements. While several self-adaptation mechanisms have been proposed in the literature, they cannot be fully applied for socio-technical systems that involve autonomous (thus, non-controllable) components. This project aims at designing and developing a runtime requirements supervision framework that monitors the execution of socio-technical systems, evaluates their behavior against the overall goals and intervenes by deciding how to revise requirements when adaptation is not possible.

@inproceedings{DBLP:journals/corr/abs-1810-10591,
  author    = {Davide Dell'Anna and
               Mehdi Dastani and
               Fabiano Dalpiaz},
  title     = {Reasoning about Norms Revision},
  booktitle   = {Preproceedings of the 29th Benelux Conference on Artificial Intelligence, BNAIC 2017},
  volume    = {abs/1810.10591},
  pages     = {281--290},
  year      = {2017},
  url_Paper       = {http://arxiv.org/abs/1810.10591},
  url_Poster= {2017_BNAIC/BNAIC17_DellAnna_Poster.pdf},
  keywords  = {Norm Revision, Norms, MAS, Multi-Agent Systems, Data Driven Supervision Of Autonomous Systems},
  archivePrefix = {arXiv},
  eprint    = {1810.10591},
  timestamp = {Sat, 23 Jan 2021 00:00:00 +0100},
  biburl    = {https://dblp.org/rec/journals/corr/abs-1810-10591.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {Norms with sanctions have been widely employed as a mechanism for controlling and coordinating the behavior of agents without limiting their autonomy. The norms enforced in a multi-agent system can be revised in order to increase the likelihood that desirable system properties are fulfilled or that system performance is sufficiently high. In this paper, we provide a preliminary analysis of some types of norm revision: relaxation and strengthening. Furthermore, with the help of some illustrative scenarios, we show the usefulness of norm revision for better satisfying the overall system objectives.}
}

Norms with sanctions have been widely employed as a mechanism for controlling and coordinating the behavior of agents without limiting their autonomy. The norms enforced in a multi-agent system can be revised in order to increase the likelihood that desirable system properties are fulfilled or that system performance is sufficiently high. In this paper, we provide a preliminary analysis of some types of norm revision: relaxation and strengthening. Furthermore, with the help of some illustrative scenarios, we show the usefulness of norm revision for better satisfying the overall system objectives.

@inproceedings{DBLP:conf/mod/AringhieriDDS17,
  author    = {Roberto Aringhieri and
               Davide Dell'Anna and
               Davide Duma and
               Michele Sonnessa},
  title     = {Evaluating the Dispatching Policies for a Regional Network of Emergency
               Departments Exploiting Health Care Big Data},
  booktitle = {Proceedings of the Third International
               Conference on Machine Learning, Optimization, and Big Data, {MOD} 2017, Revised
               Selected Papers},
  volume    = {10710},
  pages     = {549--561},
  year      = {2017},
  url_Link       = {https://doi.org/10.1007/978-3-319-72926-8_46},
  url_Paper       = {2017_MOD/MOD17_Aringhieri.pdf},
  url_Slides= {2017_MOD/MOD17_Aringhieri_Slides.pdf},
  url_Poster = {https://iris.unito.it/bitstream/2318/1624657/1/AringhieriEtAl-poster.pdf},
  keywords  = {Big Data, Health Care, Online Optimization, Big Data Health Care},
  doi       = {10.1007/978-3-319-72926-8\_46},
  timestamp = {Tue, 29 Dec 2020 00:00:00 +0100},
  biburl    = {https://dblp.org/rec/conf/mod/AringhieriDDS17.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {The Emergency Department (ED) is responsible to provide medical and surgical care to patients arriving at the hospital in need of immediate care. At the regional level, the EDs system can be seen as a network of EDs cooperating to maximise the outputs (number of patients served, average waiting time, ...) and outcomes in terms of the provided care quality. In this paper we discuss how quantitative analysis based on health care big data can provide a tool to evaluate the dispatching policies for the regional network of emergency departments: the basic idea is to exploit clusters of EDs in such a way to fairly distribute the workload. We present a simulation model based on the case study of the Piedmont in Italy. The model is powered by the knowledge provided by the analysis of the regional health care big data.}
}

The Emergency Department (ED) is responsible to provide medical and surgical care to patients arriving at the hospital in need of immediate care. At the regional level, the EDs system can be seen as a network of EDs cooperating to maximise the outputs (number of patients served, average waiting time, ...) and outcomes in terms of the provided care quality. In this paper we discuss how quantitative analysis based on health care big data can provide a tool to evaluate the dispatching policies for the regional network of emergency departments: the basic idea is to exploit clusters of EDs in such a way to fairly distribute the workload. We present a simulation model based on the case study of the Piedmont in Italy. The model is powered by the knowledge provided by the analysis of the regional health care big data.

@inproceedings{DBLP:conf/aiia/DellAnna16,
  author    = {Davide Dell'Anna},
  title     = {Numerical and Temporal Planning for a Multi-agent Team Acting in the
               Real World},
  booktitle = {Proceedings of the 3rd Italian Workshop on Artificial Intelligence
               and Robotics, AIRO@AI*IA 2016},
  series    = {{CEUR} Workshop Proceedings},
  volume    = {1834},
  pages     = {10--14},
  year      = {2016},
  url       = {http://ceur-ws.org/Vol-1834/paper2.pdf},
  url_Slides= {2016_AIRO/AIRO16_DellAnna_Slides.pdf},
  keywords  = {Planning, Numerical, Temporal, MAS, Multi-Agent, UAV, Drones, SmatF2},
  timestamp = {Wed, 12 Feb 2020 16:44:29 +0100},
  biburl    = {https://dblp.org/rec/conf/aiia/DellAnna16.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
  abstract = {This work addresses the problem of automatic planning for real world problems involving (possibly heterogeneous) teams of UAVs.
              Such problems are mainly characterized by cooperation, consumable resources and continuous numeric change, as well as concurrency, time and temporal constraints.
              These features make problems nontrivial for state-of-art planners. Difficulties are mainly due to temporal constraints, especially between different agents.
              This work reports experimental results concerning both synthetic problems and real-world multi-UAV multi-target planning scenarios.
              It shows that action-based approaches to planning, after a complex encoding process, can be successfully employed (with results comparable to other state-of-art approaches) to solve real-world problems.}
}

This work addresses the problem of automatic planning for real world problems involving (possibly heterogeneous) teams of UAVs. Such problems are mainly characterized by cooperation, consumable resources and continuous numeric change, as well as concurrency, time and temporal constraints. These features make problems nontrivial for state-of-art planners. Difficulties are mainly due to temporal constraints, especially between different agents. This work reports experimental results concerning both synthetic problems and real-world multi-UAV multi-target planning scenarios. It shows that action-based approaches to planning, after a complex encoding process, can be successfully employed (with results comparable to other state-of-art approaches) to solve real-world problems.