Leveraging Filter Correlations for Deep Model Compression

Pravendra Singh, Vinay Kumar Verma, Piyush Rai, Vinay Namboodiri; The IEEE Winter Conference on Applications of Computer Vision (WACV), 2020, pp. 835-844

Abstract


We present a filter correlation based model compression approach for deep convolutional neural networks. Our approach iteratively identifies pairs of filters with the largest pairwise correlations and drops one of the filters from each such pair. However, instead of discarding one of the filters from each such pair naively, the model is re-optimized to make the filters in these pairs maximally correlated, so that discarding one of the filters from the pair results in minimal information loss. Moreover, after discarding the filters in each round, we further finetune the model to recover from the potential small loss incurred by the compression. We evaluate our proposed approach using a comprehensive set of experiments and ablation studies. Our compression method yields state-of-the-art FLOPs compression rates on various benchmarks, such as LeNet-5, VGG-16, and ResNet-50,56, while still achieving excellent predictive performance for tasks such as object detection on benchmark datasets.

Related Material


[pdf]
[bibtex]
@InProceedings{Singh_2020_WACV,
author = {Singh, Pravendra and Verma, Vinay Kumar and Rai, Piyush and Namboodiri, Vinay},
title = {Leveraging Filter Correlations for Deep Model Compression},
booktitle = {The IEEE Winter Conference on Applications of Computer Vision (WACV)},
month = {March},
year = {2020}
}