A Framework For Pruning Deep Neural Networks Using Energy-Based Models

TL;DR

This paper introduces a flexible pruning framework for deep neural networks using energy-based models and global optimization, achieving significant parameter reduction with minimal accuracy loss.

Contribution

It presents a novel, adaptable pruning framework that leverages energy-based objectives and population-based optimization for effective neural network compression.

Findings

Over 50% parameter reduction achieved.

Less than 5% accuracy drop on CIFAR-10.

Less than 1% accuracy drop on CIFAR-100.

Abstract

A typical deep neural network (DNN) has a large number of trainable parameters. Choosing a network with proper capacity is challenging and generally a larger network with excessive capacity is trained. Pruning is an established approach to reducing the number of parameters in a DNN. In this paper, we propose a framework for pruning DNNs based on a population-based global optimization method. This framework can use any pruning objective function. As a case study, we propose a simple but efficient objective function based on the concept of energy-based models. Our experiments on ResNets, AlexNet, and SqueezeNet for the CIFAR-10 and CIFAR-100 datasets show a pruning rate of more than $50\%$ of the trainable parameters with approximately $<5\%$ and $<1\%$ drop of Top-1 and Top-5 classification accuracy, respectively.