A Highly Effective Low-Rank Compression of Deep Neural Networks with Modified Beam-Search and Modified Stable Rank

2021/11/30 by Moonjung Eo, Eo, Moonjung, Suhyun Kang +3
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Sparse and Compressive Sensing Techniques #cs.AI #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2111.15179

8 pages, 8 figures, 2 tables

openalex publication_date 2021/11/30 · arxiv created 2021/12/01 · arxiv updated 2021/12/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Compression has emerged as one of the essential deep learning research topics, especially for the edge devices that have limited computation power and storage capacity. Among the main compression techniques, low-rank compression via matrix factorization has been known to have two problems. First, an extensive tuning is required. Second, the resulting compression performance is typically not impressive. In this work, we propose a low-rank compression method that utilizes a modified beam-search for an automatic rank selection and a modified stable rank for a compression-friendly training. The resulting BSR (Beam-search and Stable Rank) algorithm requires only a single hyperparameter to be tuned for the desired compression ratio. The performance of BSR in terms of accuracy and compression ratio trade-off curve turns out to be superior to the previously known low-rank compression methods. Furthermore, BSR can perform on par with or better than the state-of-the-art structured pruning methods. As with pruning, BSR can be easily combined with quantization for an additional compression.

Citations

• Distilling the Knowledge in a Neural Network
• Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
• Fixed Point Quantization of Deep Convolutional Networks
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Trained Rank Pruning for Efficient Deep Neural Networks
• Stable Rank Normalization for Improved Generalization in Neural Networks and GANs
• Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey

Related

• Compression of Recurrent Neural Networks using Matrix Factorization
• Data-Driven Low-Rank Neural Network Compression
• Trained Rank Pruning for Efficient Deep Neural Networks
• Low Rank Optimization for Efficient Deep Learning: Making A Balance between Compact Architecture and Fast Training
• Structure-Preserving Network Compression Via Low-Rank Induced Training Through Linear Layers Composition