CISE Seminar: Mengye Ren, New York University

The Always-Learning Machine

Today’s AI models acquire most of their knowledge through offline, Independent and Identically Distributed (IID) learning. In-context learning offers some capacity for online adaptation, but a crucial question remains: can models keep learning at deployment, or even learn from scratch, through continuous streams of experience? In this talk, Mengye will present several recent efforts toward building always-learning machines for perception and planning. Starting with experiential video streams, he will show how event segmentation (clustering event concepts in lifelong video) enables effective visual representation learning and event recognition from scratch. In JEPA world models, always-learning can yield rapid test-time learning and generalization for planning. Finally, he will discuss his recent work on creative exploration, and on linking always learning and world modeling to the self.

Mengye Ren is an Assistant Professor of Computer Science and Data Science at New York University, where he runs the Agentic Learning AI Lab. Before joining NYU, he was a visiting faculty researcher at Google Brain Toronto and a senior research scientist at Uber Advanced Technologies Group (ATG) and Waabi, working on self-driving vehicles. He received his Ph.D. in Computer Science from the University of Toronto. His research focuses on making machine learning more natural and human-like, enabling AI to continually learn, adapt, and reason in naturalistic environments.

Faculty Host: Tianyu Wang

Student Host: Jiatong Guo