Compact Mixed-Signal Convolutional Neural Network Using a Single Modular Neuron
- 29 July 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 67 (12) , 5189-5199
Abstract
This paper demonstrates a compact mixed-signal (MS) convolutional neural network (CNN) design procedure by proposing a MS modular neuron unit that alleviates analog circuit related design issues such as noise. Through the first step of the proposed procedure, we design a CNN in software with a minimized number of channels for each layer while satisfying the network performance to the target, which creates low representational and computational cost. Then, the network is reconstructed and retrained with a single modular neuron that is recursively utilized for the entire network for the maximum hardware efficiency with a fixed number of parameters that consider signal attenuation. For the last step of the proposed procedure, the parameters of the networks are quantized to an implementable level of MS neurons. We designed the networks for MNIST and Cifar-10 and achieved compact CNNs with a single MS neuron with 97% accuracy for MNIST and 85% accuracy for Cifar-10 whose representational cost and computational cost are reduced to least two times smaller than prior works. The estimated energy per classification of the hardware network for Cifar-10 with a single MS neuron, designed with optimum noise and matching requirements, is 0.5μ J, which is five times smaller than its digital counterpart.Keywords
Funding Information
- Samsung Research Funding Center of Samsung Electronics (SRFC-IT1502-04)
This publication has 20 references indexed in Scilit:
- An Always-On 3.8 $\mu$ J/86% CIFAR-10 Mixed-Signal Binary CNN Processor With All Memory on Chip in 28-nm CMOSIEEE Journal of Solid-State Circuits, 2018
- A Neuromorphic Chip Optimized for Deep Learning and CMOS Technology With Time-Domain Analog and Digital Mixed-Signal ProcessingIEEE Journal of Solid-State Circuits, 2017
- In-Memory Computation of a Machine-Learning Classifier in a Standard 6T SRAM ArrayIEEE Journal of Solid-State Circuits, 2017
- Analysis and Design of a Passive Switched-Capacitor Matrix Multiplier for Approximate ComputingIEEE Journal of Solid-State Circuits, 2016
- Ultra-Low-Energy Mixed-Signal IC Implementing Encoded Neural NetworksIEEE Transactions on Circuits and Systems I: Regular Papers, 2016
- A 0.6 V 12 b 10 MS/s Low-Noise Asynchronous SAR-Assisted Time-Interleaved SAR (SATI-SAR) ADCIEEE Journal of Solid-State Circuits, 2016
- Mastering the game of Go with deep neural networks and tree searchNature, 2016
- Realizing Low-Energy Classification Systems by Implementing Matrix Multiplication Directly Within an ADCIEEE Transactions on Biomedical Circuits and Systems, 2015
- Passive charge redistribution digital‐to‐analogue multiplierElectronics Letters, 2015
- A 10b/12b 40 kS/s SAR ADC With Data-Driven Noise Reduction Achieving up to 10.1b ENOB at 2.2 fJ/Conversion-StepIEEE Journal of Solid-State Circuits, 2013