Design space exploration of neural network activation function circuits

Files
1810.08650v1.pdf(455.55 KB)
Accepted manuscript
Date
2019-10
Authors
Yang, Tao
Wei, Yadong
Tu, Zhijun
Zeng, Haolun
Kinsy, Michel A.
Zheng, Nanning
Ren, Pengju
Version
Accepted manuscript
OA Version
Citation
Tao Yang, Yadong Wei, Zhijun Tu, Haolun Zeng, Michel A Kinsy, Nanning Zheng, Pengju Ren. 2019. "Design Space Exploration of Neural Network Activation Function Circuits." IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Volume 38, Issue 10, pp. 1974 - 1978. https://doi.org/10.1109/tcad.2018.2871198
Abstract
The widespread application of artificial neural networks has prompted researchers to experiment with field-programmable gate array and customized ASIC designs to speed up their computation. These implementation efforts have generally focused on weight multiplication and signal summation operations, and less on activation functions used in these applications. Yet, efficient hardware implementations of nonlinear activation functions like exponential linear units (ELU), scaled ELU (SELU), and hyperbolic tangent (tanh), are central to designing effective neural network accelerators, since these functions require lots of resources. In this paper, we explore efficient hardware implementations of activation functions using purely combinational circuits, with a focus on two widely used nonlinear activation functions, i.e., SELU and tanh. Our experiments demonstrate that neural networks are generally insensitive to the precision of the activation function. The results also prove that the proposed combinational circuit-based approach is very efficient in terms of speed and area, with negligible accuracy loss on the MNIST, CIFAR-10, and IMAGE NET benchmarks. Synopsys design compiler synthesis results show that circuit designs for tanh and SELU can save between ×3.13∼×7.69 and ×4.45∼×8.45 area compared to the look-up table/memory-based implementations, and can operate at 5.14 GHz and 4.52 GHz using the 28-nm SVT library, respectively. The implementation is available at: https://github.com/ThomasMrY/ActivationFunctionDemo.
Description
License
Copyright notice: "© 2016 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."