Batch Normalization Tells You Which Filter is Important
This addresses efficiency in CNNs for computer vision applications, but it is incremental as it builds on existing pruning methods.
The paper tackles filter pruning in convolutional neural networks by using batch normalization parameters to estimate filter importance without training data, achieving outstanding performance on CIFAR-10 and ImageNet with reduced computational complexity and parameters.
The goal of filter pruning is to search for unimportant filters to remove in order to make convolutional neural networks (CNNs) efficient without sacrificing the performance in the process. The challenge lies in finding information that can help determine how important or relevant each filter is with respect to the final output of neural networks. In this work, we share our observation that the batch normalization (BN) parameters of pre-trained CNNs can be used to estimate the feature distribution of activation outputs, without processing of training data. Upon observation, we propose a simple yet effective filter pruning method by evaluating the importance of each filter based on the BN parameters of pre-trained CNNs. The experimental results on CIFAR-10 and ImageNet demonstrate that the proposed method can achieve outstanding performance with and without fine-tuning in terms of the trade-off between the accuracy drop and the reduction in computational complexity and number of parameters of pruned networks.