This is the second edition of the Bayes-duality workshop focusing on the development of AI that learns adaptively, robustly and continuously, like humans. The main goal of the meeting is to discuss (and refine) the long-term goals of the project and foster collaborations among the attendees. Part 1 of the workshop is open to public (through live streaming) and will feature talks, tutorials, and discussion by world-leading researchers working on related topics. Part 2 and 3 will be closed meetings exclusively for the collaborators of the Bayes-duality project.
Similarly to the previous edition, we hope that the proposed activities help attendees increase their research's impact, outreach and help them network. This workshop is supported by RIKEN-AIP subsidy budget and the JST CREST Grant JPMJCR2112 on “A new Bayes-duality principle for adaptive, robust, and lifelong learning of AI systems”.
In-person attendance is limited to RIKEN-AIP researchers and invited guests, but all talks in Part 1 (June 12-21) will be streamed live and zoom link can be obtained by registering at doorkeeper.
About Talks and Tutorials
University of Alberta, Canada
ETH, Switzerland
University of Illinois Urbana-Champaign, US
NTT Corporation, Japan
Google DeepMind, US
Apple, France
Sony Computer Science Laboratories, Japan
Seoul National University, South Korea
KAIST, South Korea
KAUST, Saudi Arabia
Google Research, US
TU Darmstadt, Germany
University of Colorado, US
TU Darmstadt, Germany
Google DeepMind, US
University of Alberta, Canada
École Polytechnique de Montréal, Canada
Harvard University, US
University of Oxford, UK
Helmholtz AI, Germany
Imperial College London, UK
Bioptimus, US
All eight days of the workshop are available in the following youtube playlist. Quick access to specific talks and tutorials is available in the detailed program.
* Due to a fire/earthquake drill from 15:30 to 16:10, the regular Break will start earlier. Last talk will begin once the drill is over, and the panel will follow a bit later than scheduled.
Each speaker can join Zoom through the link sent to them by email. This link allows you to join as a ‘co-host’. Please do not try to use other links (e.g., the one in doorkeeper). Contact Hugo Monzon, Cong Bai, or Clément Bazan, in case of zoom trouble.