Explainability and Interpretability When Using Fuzzy Sets and Systems: Opportunities and Challenges

20 July, 2020
15:00 – 17:00

Organizers (and Panelists):
Jerry M. Mendel, University of Southern California, Los Angeles, CA, USA
Hani Hagras, University of Essex, Colchester, UK

Panelists:
Jonathan M. Garibaldi, University of Nottingham, Nottingham, UK
Francisco Herrera, Granada Univ., Granada, Spain
Hisao Ishibuchi, Southern University of Science and Technology, Shenzhen, China
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland (email: janusz.kacprzyk@ibspan.waw.pl)

Abstract

Explainable AI (XAI) is today a worldwide hot research and development topic. This is due to the rapid increase in computing power coupled with a rapid growth in the availability and quantity of data. Explanations use words, or words and data, and so fuzzy sets and systems seem like natural and valuable contributors to XAI, but to-date they do not seem to be recognized as such by the XAI community. The purpose of this panel session is to address this in a constructive way. To do this we will assemble a panel of researchers from the fuzzy logic community who are already known to have published about interpretability of fuzzy sets.

Unlike most panel sessions where panelists present mini papers, this panel session will not do this. Instead, a small set of questions will be generated, one panelist will be asked to lead off the discussion for each question, after which the remaining panelists will have an opportunity to make remarks. Audience participation will be encouraged. Tentative questions are:

  • What are the unique contributions that fuzzy sets and systems can make to XAI?
  • What is meant by “interpretable fuzzy sets and systems” for academia versus for industry?
  • Do we need global or local interpretability when it comes to fuzzy systems and their decisions?
  • Does Computing With Words have a role to play in XAI?
  • What are the steps needed to allow fuzzy sets and systems to be used as a basis of future XAI when most people are not even aware of fuzzy sets and systems as valid XAI techniques?

Issues of Cutting Edge AI and Advanced Computational Intelligence Research

21 July, 2020
14:30 – 16:30

Panel Chair: Jose Principe

Panelists:
Dr. Jose Principe , U. of Florida
Dr. Donald Wunsch, Missouri University of Science and technology
Dr. Deven Desai, Georgia Institute of Technology, Scheller College of Business
Dr. Danil Prokhorov,  Toyota Motor Corporation
Dr. Zeynep Engin, University College of London

Abstract

Spectacular successes of deep neural networks produce AI with super-human performance in various practical problem domains. These achievements help to drastically increase our quality of life. At the same time, the success of AI raises many important questions and potential worries related to the future of humanity. This panel gives a forum to discuss about the ethical implications of AI, including: How Close Are We to Creating an AI Program which parallels general human intelligence? What is the responsibility of scientists conducting AI research? Can We Apply Human Ethics to AI?

Landscape of Publications in Computational Intelligence 

22 July, 2020
15:30 – 17:30

Panel Chair: Jim Keller

Panelists (tentative list):
Jonathan Garibaldi, IEEE Transactions on Fuzzy Systems
Haibo He,IEEE Transactions on Neural Networks and Learning Systems
Kay Chen Tan, IEEE Transactions on Evolutionary Computation
Yew Soon Ong, IEEE Transactions on Emerging Topics in Computational Intelligence
Yaochu Jin, IEEE Transactions on Cognitive and Developmental Systems
Julian Togelius, IEEE Transactions on Games
Hussein Abbass, IEEE Transactions on Artificial Intelligence
Chuan-Kang Ting, IEEE Computational Intelligence Magazine
Kenji Doya, INNS Journal, Neural Networks (published by Elsevier)
DeLiang Wang, INNS Journal, Neural Networks (published by Elsevier)

Abstract

Publish or perish is an aphorism referring to the major demand in academia and in advanced research organizations. The landscape of publishing undergoes drastic changes, and it is crucial for authors and publishers to address these challenges. This panel will provide the opportunity to exchange ideas on these critical issues, including the interdisciplinary nature of our research field, questions on the societal impact of intelligent technologies, academic honesty, open access publishing models, relationship between journal and conference papers, preprint repositories, repositories of data and methods, among others.

Funding Opportunities in Biologically Inspired AI and Computational Intelligence Research

23 July, 2020
15:30 – 17:30

Panel Chair:
Paul Werbos (NSF, Retired)
Robert Kozma (U Memphis)

Panelists (tentative list):
Nandini Iyer, AFOSR
Anthony Kuh, NSF
Hava Siegelmann, DARPA
Wlodek Duch, INCF, EU
Fahmida Chowdhury (NSF)
more TBD.

Abstract

This panel addresses novel avenues to support computational intelligence research. Representatives of funding agencies and leading experts in the field will describe research challenges and funding opportunities. Which cutting edge areas are at the focus of new funding initiatives? The panel will provide a forum for thorough discussions on these topics between the panelists. It is expected to have an intensive questions and answers section with the audience.

Introductory remarks by Dr Paul Werbos, Panel Co-Chair

Venue: With WCCI 2020 being held as a virtual conference, there will be a virtual experience of Glasgow, Scotland accessible through the virtual platform. We hope that everyone will have a chance to visit one of Europe’s most dynamic cultural capitals and the “World’s Friendliest City” soon in the future!