Winner-Take-All Autoencoders

TL;DR

This paper introduces winner-take-all autoencoders that enforce sparsity in learned representations, combining convolutional and autoencoder architectures to learn shift-invariant features with competitive classification results.

Contribution

It presents a novel winner-take-all method for hierarchical sparse representation learning in autoencoders, including convolutional variants with layer-wise training and spatial sparsity.

Findings

Learned deep sparse representations from multiple datasets.

Achieved competitive classification performance.

Demonstrated effectiveness of winner-take-all autoencoders.

Abstract

In this paper, we propose a winner-take-all method for learning hierarchical sparse representations in an unsupervised fashion. We first introduce fully-connected winner-take-all autoencoders which use mini-batch statistics to directly enforce a lifetime sparsity in the activations of the hidden units. We then propose the convolutional winner-take-all autoencoder which combines the benefits of convolutional architectures and autoencoders for learning shift-invariant sparse representations. We describe a way to train convolutional autoencoders layer by layer, where in addition to lifetime sparsity, a spatial sparsity within each feature map is achieved using winner-take-all activation functions. We will show that winner-take-all autoencoders can be used to to learn deep sparse representations from the MNIST, CIFAR-10, ImageNet, Street View House Numbers and Toronto Face datasets, and achieve competitive classification performance.