Sparse optimization guided pruning for neural networks

Yong Shi, Anda Tang, Lingfeng Niu, Ruizhi Zhou

Abstract: Highlights•We propose SOGP to enhance the connection between pretraining and pruning.•We propose a novel non-convex group sparse regularization GTl1 .•We construct a unified optimization model to integrate pruning and fine-tuning.•Extensive experiments validate the good pruning and accuracy performance of SOGP.