Dear Colleagues,
SLDS is pleased to announce that the October webinar will be at 2:00pm this Thursday. The topic is lifelong learning in structured environments. Hope to see you then.
Title: Lifelong Learning in Structured Environments
Speakers: Dr. Mengye Ren, Department of Computer Science, Courant Institute of Mathematical Sciences, New York University
Date and Time: October 26, 2023, 2:00 to 3:30 pm Eastern Time
Event Link: https://www.eventbrite.com/e/lifelong-learning-in-structured-environments-tickets-740957984277?aff=oddtdtcreator
Registration: Registration is not needed. Please use the Zoom link posted at the bottom of the Event Link.
Abstract: Real world agents like us learn from an online experience and never roll back. Lifelong learning is in stark contrast with standard iid learning of a single task. In lifelong learning, we need to deal with a small context window, a changing distribution over time, and a growing size of possible outputs. In this talk, I will discuss some recent works where we can leverage the structure of the environments to improve lifelong learning. An example of structure is the incremental appearance of the new classes, categories, and environments. In the vision domain, we can also consider the spatial temporal continuity between sequential video inputs. We also study an interleaving task structure, just like conducting repetitive review sessions in human learning. Deep neural networks trained in an interleaving task environment exhibit interesting properties. Lastly, I will discuss the safety aspects of lifelong learning, and present potential safety mechanisms by imposing task structures.
Presenter: Mengye Ren is an assistant professor of computer science and data science at New York University (NYU). Before joining NYU, he was a visiting faculty researcher at Google Brain Toronto working with Prof. Geoffrey Hinton. He received B.A.Sc. in Engineering Science (2015), and M.Sc. (2017) and Ph.D. (2021) in Computer Science from the University of Toronto, advised by Prof. Richard Zemel and Prof. Raquel Urtasun. From 2017 to 2021, he was also a senior research scientist at Uber Advanced Technologies Group (ATG) and Waabi, working on self-driving vehicles. His research focuses on making machine learning more natural and human-like, in order for AIs to continually learn, adapt, and reason in naturalistic environments.
------------------------------
Zhihua Su, PhD
Associate Professor
Department of Statistics
University of Florida
zhihuasu@stat.ufl.edu------------------------------