Weight Pruning via Adaptive Sparsity Loss

TL;DR

This paper introduces an adaptive sparsity loss method for efficient, budget-aware neural network pruning during training, significantly reducing model size with minimal computational overhead.

Contribution

It proposes a novel adaptive sparsity loss for end-to-end network pruning that automatically respects user-defined resource constraints.

Findings

Effective pruning on CIFAR and ImageNet datasets.

Compatible with various architectures like AlexNet, ResNets, Wide ResNets.

Achieves significant compression with minimal accuracy loss.

Abstract

Pruning neural networks has regained interest in recent years as a means to compress state-of-the-art deep neural networks and enable their deployment on resource-constrained devices. In this paper, we propose a robust compressive learning framework that efficiently prunes network parameters during training with minimal computational overhead. We incorporate fast mechanisms to prune individual layers and build upon these to automatically prune the entire network under a user-defined budget constraint. Key to our end-to-end network pruning approach is the formulation of an intuitive and easy-to-implement adaptive sparsity loss that is used to explicitly control sparsity during training, enabling efficient budget-aware optimization. Extensive experiments demonstrate the effectiveness of the proposed framework for image classification on the CIFAR and ImageNet datasets using different architectures, including AlexNet, ResNets and Wide ResNets.