Recklessly Approximate Sparse Coding

TL;DR

This paper reveals that simple soft threshold encoding techniques perform well in image classification because they approximate solutions to non-negative sparse coding, offering a computationally efficient alternative with demonstrated effectiveness.

Contribution

It establishes a mathematical connection between soft threshold features and approximate non-negative sparse coding, providing a new understanding of their success.

Findings

Soft threshold features approximate non-negative sparse coding solutions.

Variants of soft threshold features are effective on image classification benchmarks.

The approach offers a computationally efficient alternative to more complex encoding methods.

Abstract

It has recently been observed that certain extremely simple feature encoding techniques are able to achieve state of the art performance on several standard image classification benchmarks including deep belief networks, convolutional nets, factored RBMs, mcRBMs, convolutional RBMs, sparse autoencoders and several others. Moreover, these "triangle" or "soft threshold" encodings are ex- tremely efficient to compute. Several intuitive arguments have been put forward to explain this remarkable performance, yet no mathematical justification has been offered. The main result of this report is to show that these features are realized as an approximate solution to the a non-negative sparse coding problem. Using this connection we describe several variants of the soft threshold features and demonstrate their effectiveness on two image classification benchmark tasks.