Friday Lunch Seminar: Satoshi Nishida: “Brain-information integration to make AI behavior more human-like” (On-line for CiNet members only: Sign-up required)

Friday Lunch Seminar (English)

October 7, 2022
12:15 〜 13:00
(On-line)

Talk Title: Brain-information integration to make AI behavior more human-like

Satoshi Nishida
Senior Researcher, CiNet PI
Neural Information Engineering Laboratory
Center for Information and Neural Networks (CiNet)
National Institutes of Information and Communications Technology (NICT)

Host PI: Satoshi Nishida

Abstract:
Recent advantages in AI technology are remarkable. However, there are still many considerable differences in behavior between AI and humans. Such differences prohibit the realization of trustworthy AI, which lives with people in harmony. To address this issue, our research group has developed a novel framework to integrate brain information into AI via brain-response simulation. Our research results revealed the following capabilities of this framework: (1) Brain-information integration improves the performance of the existing AI methods on recognition tasks that require the estimation of human cognition and behavior in natural situations; (2) Individual differences in mental information are reflected in AI estimation. These findings suggest that our framework for brain-information integration has great potential to endow AI with human-like behavior.