HOTCAKE: Higher Order Tucker Articulated Kernels for Deeper CNN Compression

2020/02/28 by Rui Lin, Lin, Rui, Ching-Yun Ko +13
Computer Science · Engineering · Mathematics · #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Sparse and Compressive Sensing Techniques #Tensor decomposition and applications #cs.CV #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.2002.12663

6 pages, 5 figures

arxiv created 2020/02/28 · openalex publication_date 2020/02/28 · arxiv updated 2020/03/02 · openalex created_date 2020/03/06 · openalex updated_date 2026/07/28

Abstract

The emerging edge computing has promoted immense interests in compacting a neural network without sacrificing much accuracy. In this regard, low-rank tensor decomposition constitutes a powerful tool to compress convolutional neural networks (CNNs) by decomposing the 4-way kernel tensor into multi-stage smaller ones. Building on top of Tucker-2 decomposition, we propose a generalized Higher Order Tucker Articulated Kernels (HOTCAKE) scheme comprising four steps: input channel decomposition, guided Tucker rank selection, higher order Tucker decomposition and fine-tuning. By subjecting each CONV layer to HOTCAKE, a highly compressed CNN model with graceful accuracy trade-off is obtained. Experiments show HOTCAKE can compress even pre-compressed models and produce state-of-the-art lightweight networks.

Citations

• Improving neural networks by preventing co-adaptation of feature detectors
• Tensor Decompositions and Applications
• Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions
• Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
• ImageNet classification with deep convolutional neural networks
• Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
• Ternary Weight Networks
• Pruning Filters for Efficient ConvNets
• Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications
• Learning Multiple Layers of Features from Tiny Images

Related

• A Unifying Tensor View for Lightweight CNNs
• Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
• Stable Low-rank Tensor Decomposition for Compression of Convolutional\n Neural Network
• Convolutional Neural Network Compression through Generalized Kronecker Product Decomposition
• Towards Compact CNNs via Collaborative Compression