Deep Neural Network Compression: Ternary network

Optimizing Deep Convolution Neural Network with Ternarized Weights and High Accuracy

Related publication

Zhezhi He and Deliang Fan, “Simultaneously Optimizing Weight and Quantizer of Ternary Neural Network using Truncated Gaussian Approximation,” Conference on Computer Vision and Pattern Recognition (CVPR), June 16-20, 2019, Long Beach, CA, USA [pdf]

Zhezhi He, Boqing Gong, Deliang Fan, “Optimize Deep Convolutional Neural Network with Ternarized Weights and High Accuracy,” IEEE Winter Conference on Applications of Computer Vision, January 7-11, 2019, Hawaii, USA [pdf]

Method Summary

Deep convolution neural network has achieved great success in many artificial intelligence applications. However, its enormous model size and massive computation cost have become the main obstacle for deployment of such powerful algorithm in the low power and resource limited embedded systems. As the countermeasure to this problem, in this work, we propose statistical weight scaling and residual expansion methods to reduce the bit-width of the whole network weight parameters to ternary values (i.e. -1, 0, +1), with the objectives to greatly reduce model size, computation cost and accuracy degradation caused by the model compression. With about 16X model compression rate, our ternarized ResNet-32/44/56 could outperforms full-precision counterparts by 0.12%, 0.24% and 0.18% on CIFAR-10 dataset. We also test our ternarization method with AlexNet and ResNet-18 on ImageNet dataset, which both achieve the best top-1 accuracy compared to recent similar works, with the same 16X compression rate. If further incorporating our residual expansion method, compared to the full-precision counterpart, our ternarized ResNet-18 even improves the top-5 accuracy by 0.61% and merely degrades the top-1 accuracy only by 0.42% for ImageNet dataset, with 8X model compression rate. It outperforms the recent ABC-Net by 1.03% in top-1 accuracy and 1.78% in top-5 accuracy, with around 1.25X higher compression rate and more than 6X computation reduction due to the weight sparsity.