Deep SimNets

TL;DR

SimNets are a deep architecture generalizing ConvNets by using similarity and MEX operators to create more expressive feature spaces, leading to improved accuracy especially in resource-constrained scenarios.

Contribution

Introduction of SimNets, a deep architecture that extends ConvNets with similarity and MEX operators for enhanced expressiveness and adaptability.

Findings

SimNets outperform ConvNets in accuracy when resources are limited.

SimNets achieve comparable accuracy to state-of-the-art ConvNets with proper regularization.

SimNets realize diverse feature spaces, including kernel-based and learned spaces.

Abstract

We present a deep layered architecture that generalizes convolutional neural networks (ConvNets). The architecture, called SimNets, is driven by two operators: (i) a similarity function that generalizes inner-product, and (ii) a log-mean-exp function called MEX that generalizes maximum and average. The two operators applied in succession give rise to a standard neuron but in "feature space". The feature spaces realized by SimNets depend on the choice of the similarity operator. The simplest setting, which corresponds to a convolution, realizes the feature space of the Exponential kernel, while other settings realize feature spaces of more powerful kernels (Generalized Gaussian, which includes as special cases RBF and Laplacian), or even dynamically learned feature spaces (Generalized Multiple Kernel Learning). As a result, the SimNet contains a higher abstraction level compared to a traditional ConvNet. We argue that enhanced expressiveness is important when the networks are small due to run-time constraints (such as those imposed by mobile applications). Empirical evaluation validates the superior expressiveness of SimNets, showing a significant gain in accuracy over ConvNets when computational resources at run-time are limited. We also show that in large-scale settings, where computational complexity is less of a concern, the additional capacity of SimNets can be controlled with proper regularization, yielding accuracies comparable to state of the art ConvNets.