Exploring Linear Relationship in Feature Map Subspace for ConvNets Compression

Abstract

While the research on convolutional neural networks (CNNs) is progressing quickly, the real-world deployment of these models is often limited by computing resources and memory constraints. In this paper, we address this issue by proposing a novel filter pruning method to compress and accelerate CNNs. Our work is based on the linear relationship identified in different feature map subspaces via visualization of feature maps. Such linear relationship implies that the information in CNNs is redundant. Our method eliminates the redundancy in convolutional filters by applying subspace clustering to feature maps. In this way, most of the representative information in the network can be retained in each cluster. Therefore, our method provides an effective solution to filter pruning for which most existing methods directly remove filters based on simple heuristics. The proposed method is independent of the network structure, thus it can be adopted by any off-the-shelf deep learning libraries. Experiments on different networks and tasks show that our method outperforms existing techniques before fine-tuning, and achieves the state-of-the-art results after fine-tuning.

arXiv comment17 pages, 4 figures

References in corpus (3)

• Do Deep Nets Really Need to be Deep?
• Compressing Neural Networks with the Hashing Trick
• Compact Deep Convolutional Neural Networks With Coarse Pruning

Cited by in corpus (9)

• ProtoPShare: Prototype Sharing for Interpretable Image Classification and Similarity Discovery
• Filter Pruning by Switching to Neighboring CNNs with Good Attributes
• Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration
• Sauron U-Net: Simple automated redundancy elimination in medical image segmentation via filter pruning
• Filter Grafting for Deep Neural Networks
• Manipulating Identical Filter Redundancy for Efficient Pruning on Deep and Complicated CNN
• Progressive Learning Algorithm for Efficient Person Re-Identification
• Filter Grafting for Deep Neural Networks: Reason, Method, and Cultivation
• A One-step Pruning-recovery Framework for Acceleration of Convolutional Neural Networks