Structured Pruning for Efficient ConvNets via Incremental Regularization

Huan Wang, Qiming Zhang, Yuehai Wang, Lu Yu, Haoji Hu
2019 2019 International Joint Conference on Neural Networks (IJCNN)  
Parameter pruning is a promising approach for CNN compression and acceleration by eliminating redundant model parameters with tolerable performance loss. Despite its effectiveness, existing regularization-based parameter pruning methods usually drive weights towards zero with large and constant regularization factors, which neglects the fact that the expressiveness of CNNs is fragile and needs a more gentle way of regularization for the networks to adapt during pruning. To solve this problem,
more » ... propose a new regularization-based pruning method (named IncReg) to incrementally assign different regularization factors to different weight groups based on their relative importance, whose effectiveness is proved on popular CNNs compared with state-of-the-art methods.
doi:10.1109/ijcnn.2019.8852463 dblp:conf/ijcnn/WangZWYH19 fatcat:pusjdcff5nertcxcq2b3jl2sea