Improving and Evaluating Adversarial Robustness

[CFAR Rising Star Lecture Series]
Improving and Evaluating Adversarial Robustness by Dr Pang Tianyu
5 Dec 2023 | 2.00pm (Singapore Time)

In this talk, Dr Pang Tianyu will introduce his work on improving adversarial training from the perspectives of robustness-accuracy tradeoff and diffusion-based data augmentation. He will also demonstrate how to perform adversarial attacks against large vision-language models.

Dr Pang Tianyu
Research Scientist
Sea AI Lab Singapore
Dr Pang Tianyu is a Research Scientist at the Sea AI Lab in Singapore. He obtained his Ph.D. and B.S. degrees from Tsinghua University in 2022 and 2017, respectively. His research interests span the areas of machine learning, especially including trustworthy AI and generative models. 

Dr Pang has published 30+ papers at ICML/NeurIPS/ICLR and CVPR/ICCV/ECCV, while receiving over 6,000 citations. He was supported by the Microsoft Research Asia Fellowship, Baidu Scholarship, NVIDIA Pioneering Research Award, Zhong Shimo Scholarship, WAIC Rising Star Award, CAAI Outstanding Doctoral Dissertation Award, etc. More information about Dr Pang can be found at