Efficient Structured Pruning and Architecture Searching for Group Convolution

Abstract

Efficient inference of Convolutional Neural Networks is a thriving topic recently. It is desirable to achieve the maximal test accuracy under given inference budget constraints when deploying a pre-trained model. Network pruning is a commonly used technique while it may produce irregular sparse models that can hardly gain actual speed-up. Group convolution is a promising pruning target due to its regular structure; however, incorporating such structure into the pruning procedure is challenging. It is because structural constraints are hard to describe and can make pruning intractable to solve. The need for configuring group convolution architecture, i.e., the number of groups, that maximises test accuracy also increases difficulty. This paper presents an efficient method to address this challenge. We formulate group convolution pruning as finding the optimal channel permutation to impose structural constraints and solve it efficiently by heuristics. We also apply local search to exploring group configuration based on estimated pruning cost to maximise test accuracy. Compared to prior work, results show that our method produces competitive group convolution models for various tasks within a shorter pruning period and enables rapid group configuration exploration subject to inference budget constraints.

arXiv commentPublished as an ICCV'19 NEUARCH workshop paper

References in corpus (13)

• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Neural Architecture Search with Reinforcement Learning
• DARTS: Differentiable Architecture Search
• Rethinking the Value of Network Pruning
• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
• Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures
• Deep Roots: Improving CNN Efficiency with Hierarchical Filter Groups
• Exploring the Regularity of Sparse Structure in Convolutional Neural Networks
• Speed/accuracy trade-offs for modern convolutional object detectors
• PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions
• Interleaved Group Convolutions for Deep Neural Networks
• DropNeuron: Simplifying the Structure of Deep Neural Networks
• Building Efficient Deep Neural Networks with Unitary Group Convolutions