Venue: 1303 EECS
Miguel Bessa is an Associate Professor in the School of Engineering at Brown University. His research interests include computational mechanics and materials science, development of numerical methods, machine learning and optimization, multi-scale modeling of materials and structures. Miguel Bessa and his research group envision a new era for the design of materials and structures using artificial intelligence. Miguel received a PhD in Mechanical Engineering from Northwestern University in 2016 as a Fulbright scholar. After a short postdoctoral position at Caltech (2017) and a quick leap from Assistant to Associate Professor (2021) at Delft University of Technology, he joined the Solid Mechanics Group at Brown University in the Summer of 2022.
COOPERATIVE DATA-DRIVEN MODELING
The human brain is capable of learning tasks mostly without forgetting. However, deep neural networks suffer from catastrophic forgetting when learning tasks one after the other. We address this challenge considering a class-incremental learning scenario where the network sees test data without knowing its origin. We show the best results to date for the ImageNet dataset, outperforming by more than 20% the state of the art. The proposed method is also applied to learn material laws, illustrating its versatility. This strategy is believed to open new avenues for cooperation among different researchers and practitioners.
The MICDE Fall 2022 Seminar Series is open to all.
This seminar is hosted by the Michigan Institute for Computational Discovery & Engineering (MICDE). Prof. Bessa will be hosted by Prof. Krishna Garikipati, Professor of Mechanical Engineering and Mathematics and Director of MICDE.
This is an in-person event, Zoom link will only be provided upon request.
Graduate Certificate in Computational Discovery and Engineering, and MICDE fellows, please use this form to record your attendance.
Questions? Email MICDEemail@example.com