Recklessly Approximate Sparse Coding
This provides a mathematical justification for efficient feature encoding methods in image classification, which is incremental as it builds on existing observations.
The paper tackles the problem of explaining the high performance of simple feature encoding techniques like triangle or soft threshold encodings on image classification benchmarks, showing they are approximate solutions to non-negative sparse coding and demonstrating their effectiveness on two benchmark tasks.
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.