EZCrop: Energy-Zoned Channels for Robust Output Pruning

Abstract

Recent results have revealed an interesting observation in a trained convolutional neural network (CNN), namely, the rank of a feature map channel matrix remains surprisingly constant despite the input images. This has led to an effective rank-based channel pruning algorithm, yet the constant rank phenomenon remains mysterious and unexplained. This work aims at demystifying and interpreting such rank behavior from a frequency-domain perspective, which as a bonus suggests an extremely efficient Fast Fourier Transform (FFT)-based metric for measuring channel importance without explicitly computing its rank. We achieve remarkable CNN channel pruning based on this analytically sound and computationally efficient metric and adopt it for repetitive pruning to demonstrate robustness via our scheme named Energy-Zoned Channels for Robust Output Pruning (EZCrop), which shows consistently better results than other state-of-the-art channel pruning methods.

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Distilling the Knowledge in a Neural Network
• Pruning Convolutional Neural Networks for Resource Efficient Inference
• Automatically designing CNN architectures using genetic algorithm for image classification
• Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures
• Pruning Filters for Efficient ConvNets
• Spectral Representations for Convolutional Neural Networks
• The Singular Values of Convolutional Layers