Dirichlet Pruning for Neural Network Compression

TL;DR

This paper presents Dirichlet pruning, a novel post-processing method for neural network compression that uses variational inference over Dirichlet distributions to identify and remove unimportant units, achieving state-of-the-art results.

Contribution

Introduction of Dirichlet pruning, a fast, effective structured pruning technique using Dirichlet distributions and variational inference for neural network compression.

Findings

Achieves 45% and 58% compression on VGG and ResNet.

Requires only one epoch to train.

Provides interpretable features.

Abstract

We introduce Dirichlet pruning, a novel post-processing technique to transform a large neural network model into a compressed one. Dirichlet pruning is a form of structured pruning that assigns the Dirichlet distribution over each layer's channels in convolutional layers (or neurons in fully-connected layers) and estimates the parameters of the distribution over these units using variational inference. The learned distribution allows us to remove unimportant units, resulting in a compact architecture containing only crucial features for a task at hand. The number of newly introduced Dirichlet parameters is only linear in the number of channels, which allows for rapid training, requiring as little as one epoch to converge. We perform extensive experiments, in particular on larger architectures such as VGG and ResNet (45% and 58% compression rate, respectively) where our method achieves the state-of-the-art compression performance and provides interpretable features as a by-product.