Bi-level Doubly Variational Learning for Energy-based Latent Variable Models

TL;DR

This paper introduces BiDVL, a bi-level variational learning framework for energy-based latent variable models, improving training efficiency and performance in image generation, reconstruction, and out-of-distribution detection.

Contribution

It proposes a novel bi-level optimization approach with tractable variational distributions to enhance learning of deep energy-based latent variable models.

Findings

Achieves impressive image generation results

Demonstrates strong image reconstruction capabilities

Effective out-of-distribution detection

Abstract

Energy-based latent variable models (EBLVMs) are more expressive than conventional energy-based models. However, its potential on visual tasks are limited by its training process based on maximum likelihood estimate that requires sampling from two intractable distributions. In this paper, we propose Bi-level doubly variational learning (BiDVL), which is based on a new bi-level optimization framework and two tractable variational distributions to facilitate learning EBLVMs. Particularly, we lead a decoupled EBLVM consisting of a marginal energy-based distribution and a structural posterior to handle the difficulties when learning deep EBLVMs on images. By choosing a symmetric KL divergence in the lower level of our framework, a compact BiDVL for visual tasks can be obtained. Our model achieves impressive image generation performance over related works. It also demonstrates the significant capacity of testing image reconstruction and out-of-distribution detection.