EXTRAAMAS 2020

   page       attach   
2nd International Workshop on EXplainable TRansparent Autonomous Agents and Multi-Agent Systems
Auckland, New Zealand, 09/05/2020

The domain of eXplainable Artificial Intelligence (XAI) emerged to explain the often-opaque decision mechanisms of machine learning algorithms and autonomous systems. In particular, as intelligent agents and robots get more complex and more involved into the daily lives of millions of users, making agents and robots decision-making processes explainable is a chief priority to enhance their acceptability, avoid failures, and comply with national and international relevant regulations. The 2020 edition of the EXplainable TRansparent Autonomous Agents and Multi-Agent Systems (EXTRAAMAS) aims to pursue the successful track of workshops initiated last year at 2019 in Montreal. In particular, EXTRAAMAS 2020 sets the following aims i) to strengthen the common ground for the study and development of explainable and understandable autonomous agents, robots and Multi-Agent Systems (MAS), ii) to investigate the potential of agent-based systems in the development of personalized user-aware explainable AI, iii) to assess the impact of transparent and explained solutions on the user/agents behaviors, (iv) to discuss motivating examples and concrete applications in which the lack of explainability leads to problems, which would be resolved by explainability, and (v) to assess and discuss the first demonstrators and proof of concepts paving the way for the next generation systems. To be accepted in the workshop, submission will go three a rigorous peer-review process where each paper will be reviewed by at least 3 members of the PC. Accepted papers will be presented in the 2-day workshop. In addition, a keynote on XAI given by Prof. Tim Miller is scheduled.

temi di interesse

The main aim of this second “International workshop on EXplainable TRansparent Autonomous Agents and Multi-Agent Systems” (EXTRAAMAS) is four-folded:

  • to establish a common ground for the study and development of explainable and understandable autonomous agents, robots and Multi-Agent Systems (MAS),
  • to investigate the potential of agent-based systems in the development of personalized user-aware explainable AI,
  • to assess the impact of transparent and explained solutions on the user/agents behaviors, and
  • to discuss motivating examples and concrete applications in which the lack of explainability leads to problems, which would be resolved by explainability.

Contributions are encouraged in both theoretical and practical applications for transparent and explainable intelligence in agents and MAS. Papers presenting theoretical contributions, designs, prototypes, tools, subjective user tests, assessment, new or improved techniques, and general survey papers tracking current evolutions and future directions are welcome. 

evento ospitante
evento colocato
worldBlue Sky 2020@AAMAS 2020
worldEMAS 2020@AAMAS 2020
worldHIMAS 2020@AAMAS 2020
funge da
evento origine per pubblicazione
page_white_acrobatAgent-Based Explanations in AI: Towards an Abstract Framework (articolo in atti, 2020) — Giovanni Ciatto, Michael I. Schumacher, Andrea Omicini, Davide Calvaresi
evento contenitore per presentazione
page_white_powerpointAgent-Based Explanations in AI: Towards an Abstract Framework (EXTRAAMAS 2020@AAMAS 2020, 09/05/2020) — Davide Calvaresi (Giovanni Ciatto, Michael I. Schumacher, Andrea Omicini, Davide Calvaresi)
evento ospitato per
evento colocato per
worldHIMAS 2020@AAMAS 2020
worldEMAS 2020@AAMAS 2020
worldBlue Sky 2020@AAMAS 2020
evento della serie