Online & University of Alberta Campus (CSC B-10)
Presenter: J. Fernando Hernandez Garcia, PhD student, University of Alberta; supervised by Dr. Rich Sutton
Title: Maintaining Plasticity Independent of Architecture Through Selective Reinitialization
Abstract: The problem of loss of plasticity in neural networks, in which a network loses its ability to learn from new observations when trained for an extended time, is a major limitation to implementing deep learning systems that learn continually. A tried and tested idea for maintaining plasticity is to sporadically reinitialize low-utility features in the network, an algorithm known as continual backpropagation. However, measuring the utility of features depends on the feature's connectivity pattern, which makes it difficult to combine continual backpropagation with any arbitrary network. This drawback is removed if one works at the lowest level in a network: the weights. In this talk, I present the successes and failures of continual backpropagation in maintaining plasticity in different architectures. Then, I present a new algorithm, selective weight reinitialization, which successfully maintains plasticity by reinitializing weights instead of features.
Bio: J. Fernando Hernandez-Garcia is PhD student in the RLAI Lab at the University of Alberta, supervised by Dr. Richard Sutton. He aims to design intelligent systems that continually learn while interacting with the world.
About AI Seminar
Hosted by Amii, AI Seminar is a weekly meeting where students, developers, and professors in the AI field share their current research. Presenters include speakers from the University of Alberta, local industry, and other institutions from across Canada and abroad. Once a month, Technology Alberta participates as a co-host, featuring new and exciting Alberta Tech companies.
Learn more and see all upcoming seminars on the AI Seminar website.
Looking to build AI capacity? Need a speaker at your event?
