Surbhi Goel

Surbhi Goel is the Magerman Term Assistant Professor at the University of Pennsylvania's Department of Computer and Information Science, affiliated with the School of Engineering and Applied Science. She contributes to foundational machine learning research through her roles in the ASSET Center for AI systems and the Warren Center for network sciences. Her work bridges theoretical computer science and practical deep learning advancements.

• Education: PhD in Computer Science (University of Texas at Austin), Bachelors (IIT Delhi)
• Former roles: Postdoctoral researcher at Microsoft NYC, Graduate fellowships including Simons-Berkeley and JP Morgan AI

Research focuses on deep learning theory, transformer architecture analysis, and optimization mysteries, with publications exploring neural tangent kernels, in-context learning, and adversarial robustness. Recent projects examine self-supervised learning, diffusion models, and Markov chain dynamics in language models.

Key awards include the Bert Kay Dissertation Award (2020), Rising Stars in ML (2019), and multiple competitive fellowships. She leads courses like CIS 7000: Foundations of Modern ML and CIS 5200: Machine Learning, emphasizing theoretical foundations and empirical validation.