Timezone: »

 
Poster
Learning Equivariant Energy Based Models with Equivariant Stein Variational Gradient Descent
Priyank Jaini · Lars Holdijk · Max Welling

Thu Dec 09 12:30 AM -- 02:00 AM (PST) @
in Poster Session 5 »

We focus on the problem of efficient sampling and learning of probability densities by incorporating symmetries in probabilistic models. We first introduce Equivariant Stein Variational Gradient Descent algorithm -- an equivariant sampling method based on Stein's identity for sampling from densities with symmetries. Equivariant SVGD explicitly incorporates symmetry information in a density through equivariant kernels which makes the resultant sampler efficient both in terms of sample complexity and the quality of generated samples. Subsequently, we define equivariant energy based models to model invariant densities that are learned using contrastive divergence. By utilizing our equivariant SVGD for training equivariant EBMs, we propose new ways of improving and scaling up training of energy based models. We apply these equivariant energy models for modelling joint densities in regression and classification tasks for image datasets, many-body particle systems and molecular structure generation.

Keywords: Theory · Optimization · Machine Learning · Self-Supervised Learning

Author Information

Priyank Jaini (University of Amsterdam)
Lars Holdijk (University of Amsterdam)
Max Welling (University of Amsterdam / Qualcomm AI Research)

More from the Same Authors