Discovering Influential Factors in Variational Autoencoders

TL;DR

This paper investigates how mutual information can identify influential factors in variational autoencoders, improving interpretability and downstream task performance by supervising and discovering key learned representations.

Contribution

It introduces a mutual information-based indicator for influential factors in VAEs and proposes an algorithm to compute it, enhancing interpretability and supervision.

Findings

Mutual information correlates with influential factors in VAEs.

The proposed method identifies interpretable factors in datasets.

Mutual information aids in discovering emotion-related variants in DEAP.

Abstract

In the field of machine learning, it is still a critical issue to identify and supervise the learned representation without manually intervening or intuition assistance to extract useful knowledge or serve for the downstream tasks. In this work, we focus on supervising the influential factors extracted by the variational autoencoder(VAE). The VAE is proposed to learn independent low dimension representation while facing the problem that sometimes pre-set factors are ignored. We argue that the mutual information of the input and each learned factor of the representation plays a necessary indicator of discovering the influential factors. We find the VAE objective inclines to induce mutual information sparsity in factor dimension over the data intrinsic dimension and therefore result in some non-influential factors whose function on data reconstruction could be ignored. We show mutual information also influences the lower bound of the VAE's reconstruction error and downstream classification task. To make such indicator applicable, we design an algorithm for calculating the mutual information for the VAE and prove its consistency. Experimental results on MNIST, CelebA and DEAP datasets show that mutual information can help determine influential factors, of which some are interpretable and can be used to further generation and classification tasks, and help discover the variant that connects with emotion on DEAP dataset.