The case for spatial pooling in deep convolutional sparse coding

Abstract

The sparse representation framework is a popular approach due to its desirable theoretical guarantees and the use of sparse representations as feature vectors in machine learning problems. Another seemingly unrelated line of research is deep learning and, in particular, convolutional neural networks (CNNs) which perform extremely well on various machine learning benchmarks. Recently, in [1], a connection between CNNs and convolutional sparse coding (CSC) was established using a simplified CNN model. Motivated by the use of spatial pooling in practical CNN implementations, we investigate the effect of using spatial pooling in the CSC model. We show that the spatial pooling operations do not hinder the performance and can introduce additional benefits.