Wei Chen
Optimization, Speed-up, and Out-of-distribution Prediction in Deep Learning
Abstract
In this talk, I will introduce our investigations on how to make deep learning easier to optimize, faster to train, and more robust to out-of-distribution prediction. To be specific, we design a group-invariant optimization framework for ReLU neural networks; we compensate the gradient delay in asynchronized distributed training; and we improve the out-of-distribution prediction by incorporating “causal” invariance.
Our speaker
Wei Chen is a professor in the Institute of Computing Technology, Chinese Academy of Sciences (CAS). Before she joined CAS, she was a principal research manager in Microsoft Research Asia and the co-chair of MSR Asia Theory Center. Her current research interests include trustworthy AI, causal inference and causal learning, and theoretical machine learning. She obtained the PhD in Math from Chinese Academy of Sciences in 2011, advised by Professor Zhi-Ming Ma.
