Directed-Weighting Group Lasso for Eltwise Blocked CNN Pruning

TL;DR

This paper introduces Directed-Weighting Group Lasso (DWGL), a novel pruning method for eltwise blocked CNNs that enhances filter pruning efficiency and network compression while maintaining accuracy.

Contribution

The work proposes DWGL, which enforces index-wise incremental coefficients to improve filter pruning in eltwise layers, boosting compression rates in CNNs.

Findings

Achieved 75.34% compression on ResNet-56 with minimal accuracy loss.

Attained 53% compression on ResNet-50 on ImageNet, doubling inference speed.

Demonstrated effective pruning across multiple ResNet architectures.

Abstract

Eltwise layer is a commonly used structure in the multi-branch deep learning network. In a filter-wise pruning procedure, due to the specific operation of the eltwise layer, all its previous convolutional layers should vote for which filters by index should be pruned. Since only an intersection of the voted filters is pruned, the compression rate is limited. This work proposes a method called Directed-Weighting Group Lasso (DWGL), which enforces an index-wise incremental (directed) coefficient on the filterlevel group lasso items, so that the low index filters getting high activation tend to be kept while the high index ones tend to be pruned. When using DWGL, much fewer filters are retained during the voting process and the compression rate can be boosted. The paper test the proposed method on the ResNet series networks. On CIFAR-10, it achieved a 75.34% compression rate on ResNet-56 with a 0.94% error increment, and a 52.06% compression rate on ResNet-20 with a 0.72% error increment. On ImageNet, it achieved a 53% compression rate with ResNet-50 with a 0.6% error increment, speeding up the network by 2.23 times. Furthermore, it achieved a 75% compression rate on ResNet-50 with a 1.2% error increment, speeding up the network by 4 times.