Past talk · AI/ML Seminar Series
Semi-Supervised Learning with Prediction-Constrained Variational Autoencoders
PhD Student, Computer Science, University of California, Irvine
- Date & time
- Monday, May 22, 2023 · 1:00 PM
- Location
- Donald Bren Hall 4011
Abstract
Variational autoencoders (VAEs) have proven to be an effective approach to modeling complex data distributions while providing compact representations useful for downstream prediction tasks. In this work we train VAEs with the dual goals of good likelihood-based generative modeling and good discriminative performance in supervised and semi-supervised prediction tasks. We show that the dominant approach to training semi-supervised VAEs has key weaknesses, and propose a novel framework that maximizes generative likelihood subject to prediction quality constraints. To handle sparse labels, we further enforce a consistency constraint requiring predictions on reconstructed data to match those on the original data. Our experiments show that prediction and consistency constraints improve generative samples as well as image classification performance in semi-supervised settings.
About the speaker
Gabe Hope is a final-year PhD student at UC Irvine working with professor Erik Sudderth. His research focuses on deep generative models, interpretable machine learning and semi-supervised learning. He will join the faculty at Harvey Mudd College as a visiting assistant professor in computer science.