skip to content

Department of Computer Science and Technology

  • PhD Student
  • MD
  • This person is no longer active in the department.

Biography

Francesco Sovrano is a PhD student in Data Science at the University of Bologna. 

His research addresses explainable and explanatory artificial intelligence (XAI & YAI) in the field of technical communication, legal data analysis (i.e. legal question answering) and digital humanities. 

More in detail, his main research project is about understanding how to formalise and automatically generate pragmatic (agent-centred) explanations of complex data and processes (e.g. XAI, LOD), with particular attention to the EU's Right to Explanation. He is developing mechanisms for explaining to humans (by exploiting Achinstein's theory of explanations) and also to machines (Reinforcement Learning agents to which explain rule-based regulations; multiple RL-agents that have to optimally explain each others).

In addition to research, Francesco also carries out teaching and tutoring activities at the University of Bologna for the department of Computer Science (on USABILITY & USER EXPERIENCE DESIGN, WEB TECHNOLOGIES) and the department of Law (on Data Science for Lawyers, Sem. Data Science For Legal Analytics).

During his PhD, Francesco has created a vivid network of collaborations resulting from visiting periods in US and UK at:

  • the University of Pittsburgh (PA; USA), hosted by Prof. Kevin Ashley and Prof. Peter Brusilovsky. Topic: Adaptive e-Learning in Law.
  • the University of Cambridge (UK), hosted by Prof. Amanda Prorok. Topic: Explanation-Aware Multi-Agent Reinforcement Learning. 
  • the Stanford University (CA; US) for the European Project Marie Curie 'MIREL-MInig and REasoning with Legal texts '. Topic: automatic extraction of legal ontology design patterns starting from legal texts.

Francesco also participated to several European projects:

  • the MIREL (MIREL: MIning and REasoning with Legal texts) project, funded by the Justice program of the European Union under the Grant Agreement n. 690974.
  • the InterLex project, funded by the Justice program of the European Union under the Grant Agreement No.800839.
  • the LEOS (Legislation Editing Open Software) project.

 

Research

Francesco studies how to formalise and automatically generate pragmatic (agent-centred) explanations of complex data and processes (e.g. XAI, LOD), with particular attention to the EU's Right to Explanation. I am developing mechanisms for explaining to humans (by exploiting Achinstein's theory of explanations) and also to machines (Reinforcement Learning agents to which explain rule-based regulations; multiple RL-agents that have to optimally explain each others).

Professional Activities

Prizes & Awards

 

     Education & Training

    • (2018 - 2022) PhD in Data Science & Computation; Università di Bologna - Department of Computer Science and Engineering
    • (2016 - 2018) Master Degree in Computer Science; Università di Bologna

     

    Work experience

    • (2022/03) Alma Human AI Centre of the University of Bologna: OCCASIONAL SELF-EMPLOYMENT CONTRACT; development of AI models applied to European legislation with the goal of defining best practices to encourage legal drafting activities and support "better regulation".
    • (2021/12) Contract on the European project LEOS - Legislation Editing Open Software.
    • (2021/09 - 2022/03) Tutoring activities related to the teaching of WEB TECHNOLOGIES [cod. 41731]. Academic Year 2021/2022 for the COMPUTER SCIENCE degree.
    • (2021/09 - 2021/12) Teaching activity. 8 hours of lessons held as part of the 'USABILITY & USER EXPERIENCE DESIGN' course, on explainable and explanatory artificial intelligence (XAI & YAI). ID-90720 (https://www.unibo.it/en/teaching/course-unit-catalogue/course-unit/2021/...). 
    • (2021/09 - 2021/11) Teaching activity. 10 hours of lessons held as part of the 'Data Science for Lawyers' course, on data analysis and visualisation and how to lie with data. ID-96482 (https://www.unibo.it/en/teaching/course-unit-catalogue/course-unit/2021/...).
    • (2019/07 - 2021/10) Research activity within the MIREL project, funded by the Justice program of the European Union under the Grant Agreement n. 690974. “MIREL: MIning and REasoning with Legal texts” - Grant Agreement Number 690974; research activities summarized in the following scientific publications: http://dx.doi.org/10.1007/978-3-030-89811-3_12 ; http://dx.doi.org/10.3233/FAIA190326
    • (2019/09 - 2021/09) InterLex - Grant Agreement No.800839: Research activity within the InterLex project, funded by the Justice program of the European Union under the Grant Agreement n. 800839.
    • (2021/03 - 2021/05) Teaching activity. 8 hours of lessons held as part of the 'Sem. Data Science For Legal Analytic' course, on explainable and explanatory artificial intelligence (XAI & YAI). ID-94376 (https://www.unibo.it/en/teaching/course-unit-catalogue/course-unit/2020/...).
    • (2020/09 - 2021/03) Tutoring activities related to the teaching of WEB TECHNOLOGIES [cod. 41731]. Academic Year 2020/2021 for the COMPUTER SCIENCE degree.
    • (2020/09 - 2020/12) Teaching activity. 8 hours of lessons held as part of the 'USABILITY & USER EXPERIENCE DESIGN' course, on explainable and explanatory artificial intelligence (XAI & YAI). ID-90720 (https://www.unibo.it/en/teaching/course-unit-catalogue/course-unit/2020/...).
    • (2020/04) Consultant for CINI - Consorzio Interuniversitario Nazionale per l'Informatica: Applications in the legal field of Artificial Intelligence.
    • (2019/09) Teaching assignment in the LEX Summer School 2019.
    • (2019/09 - 2020/03) Tutoring activities related to the teaching of WEB TECHNOLOGIES [cod. 41731]. Academic Year 2019/2020 for the COMPUTER SCIENCE degree.
    • (2018/09 - 2019/03) Tutoring activities related to the teaching of PROGRAMMING [cod. 00819]. Academic Year 2018/2019 for the COMPUTER SCIENCE degree.

     

    Hobby & Open-Source Software

     

    Publications

    • Sovrano, Francesco; Raymond, Alex; Prorok, Amanda, Explanation-Aware Experience Replay in Rule-Dense Environments, «IEEE ROBOTICS AND AUTOMATION LETTERS», 2022, 7, pp. 898 - 905 [Scientific article]
    • Sovrano, Francesco; Sapienza, Salvatore; Palmirani, Monica; Vitali, Fabio, Metrics, Explainability and the European AI Act Proposal, «J», 2022, 5, pp. 126 - 138 [Scientific article]
    • Sovrano, Francesco; Palmirani, Monica; Distefano, Biagio; Sapienza, Salvatore; Vitali, Fabio, A dataset for evaluating legal question answering on private international law, in: Proceedings of the Eighteenth International Conference on Artificial Intelligence and Law, New York, Association for Computing Machinery, 2021, pp. 230 - 234 (acts of: International Conference on Artificial Intelligence and Law, São Paulo, Brazil, June 21 - 25, 2021) [Contribution to conference proceedings]Open Access
    • Sovrano, Francesco; Sapienza, Salvatore; Palmirani, Monica; Vitali, Fabio, A Survey on Methods and Metrics for the Assessment of Explainability Under the Proposed AI Act, in: Legal Knowledge and Information Systems, 2021, pp. 235 - 242 (acts of: JURIX 2021: 4th International Conference on Legal Knowledge and Information Systems, Mykolas Romeris University, Vilnius, Lithuania, December 8-10, 2021) [Contribution to conference proceedings]
    • Sovrano, Francesco; Vitali, Fabio, From Philosophy to Interfaces: an Explanatory Method and a Tool Inspired by Achinstein’s Theory of Explanation, in: IUI '21: 26th International Conference on Intelligent User Interfaces, 2021, pp. 81 - 91 (acts of: IUI '21: 26th International Conference on Intelligent User Interfaces, College Station TX USA, April, 2021) [Contribution to conference proceedings]
    • Palmirani, Monica; Sovrano, Francesco; Liga, Davide; Sapienza, Salvatore; Vitali, Fabio, Hybrid AI Framework for Legal Analysis of the EU Legislation Corrigenda, in: Legal Knowledge and Information Systems, 2021, pp. 68 - 75 (acts of: JURIX 2021: 34th International Conference on Legal Knowledge and Information Systems, Mykolas Romeris University, Vilnius, Lithuania, December 8-10, 2021) [Contribution to conference proceedings]
    • Sovrano, Francesco; Vitali, Fabio; Palmirani, Monica, Making Things Explainable vs Explaining: Requirements and Challenges Under the GDPR, in: AI Approaches to the Complexity of Legal Systems XI-XII, Madrid, Víctor Rodríguez-Doncel, Monica Palmirani, Michał Araszkiewicz, Pompeu Casanovas, Ugo Pagallo, Giovanni Sartor, 2021, pp. 169 - 182 (LECTURE NOTES IN ARTIFICIAL INTELLIGENCE) [Chapter or essay]
    • Asperti, Andrea; Cortesi, Daniele; De Pieri, Carlo; Pedrini, Gianmaria; Sovrano, Francesco, Crawling in Rogue's Dungeons With Deep Reinforcement Techniques, «IEEE TRANSACTIONS ON GAMES», 2020, 12, pp. 177 - 186 [Scientific article]
    • Sovrano, Francesco; Palmirani, Monica; Vitali, Fabio, Deep learning based multi-label text classification of UNGA resolutions, in: ICEGOV 2020: Proceedings of the 13th International Conference on Theory and Practice of Electronic Governance, New York, ACM Association for Computing Machinery, 2020, pp. 686 - 695 (acts of: ICEGOV 2020: International Conference on Theory and Practice of Electronic Governance, Athen, 23-25 September 2020) [Contribution to conference proceedings]Open Access
    • Palmirani, Monica; Bincoletto, Giorgia; Leone, Valentina; Sapienza, Salvatore; Sovrano, Francesco, Hybrid Refining Approach of PrOnto Ontology, in: Electronic Government and the Information Systems Perspective., Cham, Springer, 2020, 12394, pp. 3 - 17 (acts of: 9th International Conference on Electronic Government and the Information Systems Perspective. EGOVIS 2020, Bratislava, Slovakia, September 14 - 17, 2020) [Contribution to conference proceedings]
    • Francesco Sovrano, Monica Palmirani, Fabio Vitali, Legal Knowledge Extraction for Knowledge Graph Based Question-Answering, in: Legal Knowledge and Information Systems JURIX 2020: The Thirty-third Annual Conference, Brno, Czech Republic, December 9–11, 2020, Amsterdam, IOS, 2020, 334, pp. 143 - 153 (acts of: Legal Knowledge and Information Systems (JURIX 2020), 33rd International Conference., Brno, Czech Republic, December 9–11, 2020) [Contribution to conference proceedings]Open Access
    • Sovrano, Francesco; Vitali, Fabio; Palmirani, Monica, Modelling GDPR-Compliant Explanations for Trustworthy AI, in: Electronic Government and the Information Systems Perspective, Cham, Springer, 2020, 12394, pp. 219 - 233 (acts of: International Conference on Electronic Government and the Information Systems Perspective. EGOVIS 2020, Bratislava, September 14 - 17, 2020) [Contribution to conference proceedings]
    • Sovrano, Francesco, Combining Experience Replay with Exploration by Random Network Distillation, in: 2019 IEEE Conference on Games (CoG), 2019, pp. 1 - 8 (acts of: 2019 IEEE Conference on Games (CoG), London, UK, 20-23 August) [Contribution to conference proceedings]
    • Asperti, Andrea; Cortesi, Daniele; Sovrano, Francesco, Crawling in Rogue’s Dungeons with (Partitioned) A3C, in: Machine Learning, Optimization, and Data Science - 4th International Conference, {LOD} 2018, Volterra, Italy, September 13-16, 2018, Revised Selected Papers, Springer, «LECTURE NOTES IN ARTIFICIAL INTELLIGENCE», 2019, 11331, pp. 264 - 275 (acts of: Machine Learning, Optimization, and Data Science - 4th International Conference, LOD 2018, Volterra, Italy, September 13-16, 2018) [Contribution to conference proceedings]
    • Francesco Sovrano, Fabio Vitali, Monica Palmirani, The difference between Explainable and Explaining: requirements and challenges under the GDPR, in: XAILA 2019 EXplainable AI in Law 2019. Proceedings of the 2nd EXplainable AI in Law Workshop (XAILA 2019) co-located with 32nd International Conference on Legal Knowledge and Information Systems (JURIX 2019), Aachen, CEUR-WS.org, 2019, 2681, pp. 1 - 11 (acts of: 2nd EXplainable AI in Law Workshop (XAILA 2019) co-located with 32nd International Conference on Legal Knowledge and Information Systems (JURIX 2019), Madrid, Spain, December 11, 2019) [Contribution to conference proceedings]Open Access
    • Asperti, Andrea; Carlo De Pieri, ; Maldini, Mattia; Gianmaria, Pedrini; Francesco, Sovrano, A Modular Deep-learning Environment for Rogue, «WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS», 2017, 12, pp. 362 - 373 [Scientific article]

     

    Contact Details

    Email: 

    francesco.sovrano2@unibo.it