معرفی
Eric Wong is an Assistant Professor in the Department of Computer and Information Science at the University of Pennsylvania, where he leads the Brachio Lab focused on debugging machine learning systems. He is affiliated with the ASSET Center for safe, explainable, and trustworthy AI systems. His research bridges machine learning and optimization to ensure reliability and trustworthiness in AI systems.
- Education: PhD in Computer Science from Carnegie Mellon University under Zico Kolter, postdoc with Aleksander Madry at MIT.
His research interests center on trustworthy AI, including robustness against adversarial attacks, explainability in model predictions, and data efficiency in training. Recent work explores neuro-symbolic learning, adversarial prompting, and certified defenses for large language models.
The 15 most recent publications reveal trends in LLM security, neuro-symbolic frameworks, and certified explanations. Key subfields include adversarial robustness, model unlearning, influence functions, and rule-based inference.
- Scientific Awards: Amazon Research Award (2024), Siebel Scholar Fellowship (2020), NeurIPS Best Defense Paper (2017).
As a mentor, he advises PhD students including Weiqiu You and Helen Jin, and encourages Penn students to take CIS 5200 Machine Learning and CIS 3333 Mathematics for Machine Learning for research opportunities.
