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

03/24/2022
by   Ge Kan, et al.
0

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.

READ FULL TEXT

page 7

page 13

research
10/15/2020

Bi-level Score Matching for Learning Energy-based Latent Variable Models

Score matching (SM) provides a compelling approach to learn energy-based...
research
10/16/2020

Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models

The learning and evaluation of energy-based latent variable models (EBLV...
research
07/27/2019

Variational f-divergence Minimization

Probabilistic models are often trained by maximum likelihood, which corr...
research
09/15/2022

Langevin Autoencoders for Learning Deep Latent Variable Models

Markov chain Monte Carlo (MCMC), such as Langevin dynamics, is valid for...
research
02/13/2023

GFlowNet-EM for learning compositional latent variable models

Latent variable models (LVMs) with discrete compositional latents are an...
research
05/19/2021

Patchwise Generative ConvNet: Training Energy-Based Models from a Single Natural Image for Internal Learning

Exploiting internal statistics of a single natural image has long been r...
research
07/04/2017

Learning Deep Energy Models: Contrastive Divergence vs. Amortized MLE

We propose a number of new algorithms for learning deep energy models an...

Please sign up or login with your details

Forgot password? Click here to reset