LEARNING PROTEIN FAMILY MANIFOLDS WITH SMOOTHED ENERGY-BASED MODELS
Nathan Frey ⋅ Dan Berenberg ⋅ Joseph Kleinhenz ⋅ Stephen Ra ⋅ Isidro Hotzel ⋅ Julien Lafrance-Vanasse ⋅ Ryan Kelly ⋅ Yan Wu ⋅ Arvind Rajpal ⋅ Richard Bonneau ⋅ Kyunghyun Cho ⋅ Andreas Loukas ⋅ Vladimir Gligorijevic ⋅ Saeed Saremi
2023 Poster
in
Workshop: Machine Learning for Drug Discovery (MLDD)
in
Workshop: Machine Learning for Drug Discovery (MLDD)
Abstract
We resolve difficulties in training and sampling from discrete energy-based models (EBMs) by learning a smoothed energy landscape, sampling the smoothed data manifold with Langevin Markov chain Monte Carlo, and projecting back to the true data manifold with one-step denoising. Our formalism combines the attractive properties of EBMs and improved sample quality of score-based models, while simplifying training and sampling by requiring only a single noise scale. We demonstrate the robustness of our approach on generative modeling of antibody proteins.
Video
Chat is not available.
Successful Page Load