Denoising Criterion for Variational Auto-Encoding Framework

TL;DR

This paper introduces a denoising criterion for variational autoencoders that involves injecting noise into both input and hidden layers, leading to improved training and better likelihood performance on benchmark datasets.

Contribution

It proposes a novel denoising variational autoencoder with a modified lower bound that handles input noise tractably, enhancing the VAE framework.

Findings

DVAE outperforms VAE and IWAE in average log-likelihood on MNIST.

The proposed method effectively handles corrupted inputs during training.

Experimental results demonstrate improved generative performance.

Abstract

Denoising autoencoders (DAE) are trained to reconstruct their clean inputs with noise injected at the input level, while variational autoencoders (VAE) are trained with noise injected in their stochastic hidden layer, with a regularizer that encourages this noise injection. In this paper, we show that injecting noise both in input and in the stochastic hidden layer can be advantageous and we propose a modified variational lower bound as an improved objective function in this setup. When input is corrupted, then the standard VAE lower bound involves marginalizing the encoder conditional distribution over the input noise, which makes the training criterion intractable. Instead, we propose a modified training criterion which corresponds to a tractable bound when input is corrupted. Experimentally, we find that the proposed denoising variational autoencoder (DVAE) yields better average log-likelihood than the VAE and the importance weighted autoencoder on the MNIST and Frey Face datasets.