MXQN:Mixed quantization for reducing bit-width of weights and activations in deep convolutional neural networks
- 4 January 2021
- journal article
- research article
- Published by Springer Nature in Applied Intelligence
- Vol. 51 (7) , 4561-4574
Abstract
No abstract availableKeywords
Funding Information
- National Key Research & Development Program of China (No. 2018YFB1003304)
This publication has 29 references indexed in Scilit:
- Memristive Spiking Neural Networks Trained with Unsupervised STDPElectronics, 2018
- Intuitionistic high-order fuzzy time series forecasting method based on pi-sigma artificial neural networks trained by artificial bee colonyGranular Computing, 2018
- LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks2018
- Optimization of type-1, interval type-2 and general type-2 fuzzy inference systems using a hierarchical genetic algorithm for modular granular neural networksGranular Computing, 2018
- Efficient Processing of Deep Neural Networks: A Tutorial and SurveyProceedings of the IEEE, 2017
- Weighted-Entropy-Based Quantization for Deep Neural Networks2017
- LogNet: Energy-efficient neural networks using logarithmic computation2017
- XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks2016
- Deep Residual Learning for Image Recognition2016
- 4.6 A 65nm CMOS 6.4-to-29.2pJ/FLOP@0.8V shared logarithmic floating point unit for acceleration of nonlinear function kernels in a tightly coupled processor cluster2016