[AllUsers-ISR] (TUE 13-Oct) Invited Talk [Palestra]: Prof. Chun-Shu Wei "Beyond Accuracy: Deep Learning for Robust, Interpretable, and Adaptive Brain-Computer Interfaces"

[Joao Ruivo Paulo]

Bom dia, 



Gostava de vos convidar para uma talk do Prof, Chun-Shu Wei, National Yang Ming Chiao Tung University, Taiwan, com o título “ Beyond Accuracy: Deep Learning for Robust, In ter pretable, and Adaptive Brain-Compu ter In ter faces”. 



Qua ndo: 13-Outubro (3ªF) 

Hora: com início às 14h30 

Local: Auditório ISR, piso 0 DEEC [ https://conference.cs.cityu.edu.hk/icist/plenary/plenary_Carreira.htm ] 
Talk title: 
Beyond Accuracy: Deep Learning for Robust, In ter pretable, and Adaptive Brain-Compu ter In ter faces 

Abstract: 
Deep learning has substantially advanced EEG-based brain-compu ter in ter faces (BCIs), but high decoding accuracy alone is not sufficient for reliable neurotechnology. EEG signals vary considerably across individuals, sessions, and recording conditions, while deep learning models often remain difficult to in ter pret and generalize. In this talk, I will present recent work addressing three challenges beyond accuracy. First, I will discuss robust EEG decoding methods for handling variability across subjects, dom ains, and sensing configurations. Second, I will present explainable AI approaches for understanding what deep neural networks learn from EEG and assessing the faithfulness of model explanations. Finally, I will discuss adaptive BCIs in which both the machine learning model and the human user learn through in ter action. Together, these studies aim to advance BCIs toward systems that are robust, in ter pretable, and adaptive in real-world settings. 

Short bio: 
Chun-Shu Wei is an Associate Professor of Compu ter Science at National Yang Ming Chiao Tung University, Taiwan, with a joint appointment in Biomedical Engineering, where he leads the Brain and Computational Intelligence Lab. He received his PhD in Bioengineering from the University of California San Diego and completed postdoctoral training at Stanford University. His research focuses on brain-compu ter in ter faces, deep learning, explainable AI, and closed-loop neuromodulation, with the goal of developing robust and adaptive neurotechnology for real-world and clinical applications. He serves as an Associate Editor for Frontiers in Computational Neuroscience and an Academic Editor for PLOS ONE, and is a member of the IEEE Technical Committee on Brain-Machine In ter faces. He received the U.S. National Academy of Medicine Healthy Longevity Global Grand Challenge Catalyst Award in 2024 and is a Fellow of the UK Higher Education Academy. 



João Ruivo Paulo 

Assistant Professor | Senior Researcher 





tel: +351 926810036 

Institute of Systems and Robotics 
Department of Electrical and Compu ter Engineering - University of Coimbra 


Pinhal de Marrocos - Polo II 3030 Coimbra - Portugal 
[ http://isr.uc.pt/ | http://isr.uc.pt/ ] 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isr.uc.pt/archive/allusers/attachments/20261005/54b015d5/attachment-0001.html>