Cascaded Mixed-Precision Networks

Cascaded Mixed-Precision Networks
Title:
Cascaded Mixed-Precision Networks
Other Titles:
2020 IEEE International Conference on Image Processing (ICIP)
Publication Date:
30 September 2020
Citation:
X. Geng, J. Lin and S. Li, "Cascaded Mixed-Precision Networks," 2020 IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates, 2020, pp. 241-245, doi: 10.1109/ICIP40778.2020.9190760.
Abstract:
There has been a vast literature on Neural Network Compression, either by quantizing network variables to low precision numbers or pruning redundant connections from the network architecture. However, these techniques experience performance degradation when the compression ratio is increased to an extreme extent. In this paper, we propose Cascaded Mixed-precision Networks (CMNs), which are compact yet efficient neural networks without incurring performance drop. CMN is designed as a cascade framework by concatenating a group of neural networks with sequentially increased bitwidth. The execution flow of CMN is conditional on the difficulty of input samples, i.e., easy examples will be correctly classified by going through extremely low-bitwidth networks, and hard examples will be handled by high-bitwidth networks, so that the average compute is reduced. In addition, weight pruning is incorporated into the cascaded framework and jointly optimized with the mixed-precision quantization. To validate this method, we implemented a 2-stage CMN consisting of a binary neural network and a multi-bit (e.g. 8 bits) neural network. Empirical results on CIFAR-100 and ImageNet demonstrate that CMN performs better than state-of-the-art methods, in terms of accuracy and compute.
License type:
PublisherCopyrights
Funding Info:
I2R/18-817442-R20A. This research is supported by the Agency for Science, Technology and Research (A*STAR) under its AME Programmatic Funds (Project No.A1892b0026).
Description:
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
2381-8549
1522-4880
ISBN:
978-1-7281-6395-6
978-1-7281-6394-9
978-1-7281-6396-3
Files uploaded:

File Size Format Action
icip2020-cascaded-mixed-precision-networks.pdf 243.99 KB PDF Request a copy