Presentation | 2022-12-16 Evaluation of Fault Tolerance in Stochastic Computing Based Neural Network Inference Accelerators Ryoga Nagashima, Stefan Holst, Xiaoqing Wen, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In recent years, neural networks have become increasingly complex. Stochastic computing (SC) techniques are currently being explored to address the increased circuit area and power consumption. Besides area and power benefits of SC, fault tolerance is often mentioned as a big advantage over classic binary computing. However, the effects of faults in SC have yet to be fully investigated. In this study, we focus on the fault tolerance of stochastic computing and implement the processing element (PE) of a neural network using a conventional binary circuit and a stochastic computing circuit, and conduct a comparative evaluation of the fault-tolerant performance. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Stochastic / Neural Network / Fault |
Paper # | DC2022-74 |
Date of Issue | 2022-12-09 (DC) |
Conference Information | |
Committee | DC |
---|---|
Conference Date | 2022/12/16(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Safety, etc. |
Chair | Tatsuhiro Tsuchiya(Osaka Univ.) |
Vice Chair | Toshinori Hosokawa(Nihon Univ.) |
Secretary | Toshinori Hosokawa(Nihon Univ.) |
Assistant |
Paper Information | |
Registration To | Technical Committee on Dependable Computing |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Evaluation of Fault Tolerance in Stochastic Computing Based Neural Network Inference Accelerators |
Sub Title (in English) | |
Keyword(1) | Stochastic |
Keyword(2) | Neural Network |
Keyword(3) | Fault |
1st Author's Name | Ryoga Nagashima |
1st Author's Affiliation | Kyushu Institute of Technology(Kyutech) |
2nd Author's Name | Stefan Holst |
2nd Author's Affiliation | Kyushu Institute of Technology(Kyutech) |
3rd Author's Name | Xiaoqing Wen |
3rd Author's Affiliation | Kyushu Institute of Technology(Kyutech) |
Date | 2022-12-16 |
Paper # | DC2022-74 |
Volume (vol) | vol.122 |
Number (no) | DC-318 |
Page | pp.pp.12-16(DC), |
#Pages | 5 |
Date of Issue | 2022-12-09 (DC) |