Abstract: Adversarial training often secures models against one attack type but sacrifices generalization to multiple types of attacks and clean accuracy. This talk introduces a population-based perspective for building adversarially resilient deep models. I will first present Efficient Robust Mode Connectivity (ERMC), which connects models robust to different Lp attacks through a continuous low-loss path, achieving unified multi-norm robustness. Then I will discuss the Dual-Model Mixture-of-Experts, which combines clean and robust experts within an architecture to balance accuracy and robustness. Together, these works show that population diversity, across attacks and architectures, enables collective robustness beyond what any single model can achieve.
Bio: Ren Wang is an Assistant Professor in the Department of Electrical and Computer Engineering at the Illinois Institute of Technology and a faculty member of the Institute for Data, Econometrics, Algorithms, and Learning. His research focuses on Trustworthy Machine Learning, Population-Based Machine Learning, and applications of ML in smart grids, biology, and healthcare. His long-term vision is to develop next-generation trustworthy and intelligent machine learning systems that accelerate progress in engineering and scientific discovery. Dr. Wang has published widely in top-tier conferences and journals spanning machine learning, signal processing, computer vision, power systems, and bioinformatics, and has served as an area chair for several premier conferences. He received the 2023 ORAU Ralph E. Powe Junior Faculty Enhancement Award and, as Principal Investigator, has led multiple research projects supported by U.S. federal agencies such as the NSF and DoE.