MemSE: Fast MSE Prediction for Noisy Memristor-Based DNN Accelerators

被引:0
|
作者
Kern, Jonathan [1 ,2 ]
Henwood, Sebastien [1 ]
Mordido, Goncalo [1 ,3 ]
Dupraz, Elsa [2 ]
Aissa-El-Bey, Abdeldjalil [2 ]
Savaria, Yvon [1 ]
Leduc-Primeau, Francois [1 ]
机构
[1] Polytech Montreal, Dept Elect Engn, Montreal, PQ, Canada
[2] IMT Atlantique, Lab STICC, CNRS UMR 6285, Brest, France
[3] Mila Quebec AI Inst, Montreal, PQ, Canada
关键词
OPTIMIZATION;
D O I
10.1109/AICAS54282.2022.9869978
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Memristors enable the computation of matrix-vector multiplications (MVM) in memory and, therefore, show great potential in highly increasing the energy efficiency of deep neural network (DNN) inference accelerators. However, computations in memristors suffer from hardware non-idealities and are subject to different sources of noise that may negatively impact system performance. In this work, we theoretically analyze the mean squared error of DNNs that use memristor crossbars to compute MVM. We take into account both the quantization noise, due to the necessity of reducing the DNN model size, and the programming noise, stemming from the variability during the programming of the memristance value. Simulations on pre-trained DNN models showcase the accuracy of the analytical prediction. Furthermore the proposed method is almost two order of magnitude faster than Monte-Carlo simulation, thus making it possible to optimize the implementation parameters to achieve minimal error for a given power constraint.
引用
收藏
页码:62 / 65
页数:4
相关论文
共 28 条
  • [1] A Heterogeneous Computing System with Memristor-Based Neuromorphic Accelerators
    Liu, Xiaoxiao
    Mao, Mengjie
    Li, Hai
    Chen, Yiran
    Jiang, Hao
    Yang, J. Joshua
    Wu, Qing
    Barnell, Mark
    2014 IEEE HIGH PERFORMANCE EXTREME COMPUTING CONFERENCE (HPEC), 2014,
  • [2] Efficient Identification of Critical Faults in Memristor-Based Inferencing Accelerators
    Chen, Ching-Yuan
    Chakrabarty, Krishnendu
    IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2022, 41 (07) : 2301 - 2314
  • [3] Generalize or Die: Operating Systems Support for Memristor-based Accelerators
    Bruel, Pedro
    Chalamalasetti, Sai Rahul
    Dalton, Chris
    El Hajj, Izzat
    Goldman, Alfredo
    Graves, Catherine
    Hwu, Wen-mei
    Laplante, Phil
    Milojicic, Dejan
    Ndu, Geoffrey
    Strachan, John Paul
    2017 IEEE INTERNATIONAL CONFERENCE ON REBOOTING COMPUTING (ICRC), 2017, : 89 - 96
  • [4] Harmonica: A Framework of Heterogeneous Computing Systems With Memristor-Based Neuromorphic Computing Accelerators
    Liu, Xiaoxiao
    Mao, Mengjie
    Liu, Beiye
    Li, Boxun
    Wang, Yu
    Jiang, Hao
    Barnell, Mark
    Wu, Qing
    Yang, Jianhua
    Li, Hai
    Chen, Yiran
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2016, 63 (05) : 617 - 628
  • [5] Pruning of Deep Neural Networks for Fault-Tolerant Memristor-based Accelerators
    Chen, Ching-Yuan
    Chakrabarty, Krishnendu
    2021 58TH ACM/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2021, : 889 - 894
  • [6] Enhanced memristor-based MNNs performance on noisy dataset resulting from memristive stochasticity
    Wu, Kechuan
    Wang, Xiaoping
    IET CIRCUITS DEVICES & SYSTEMS, 2019, 13 (05) : 704 - 709
  • [7] An Ultra-Efficient Memristor-Based DNN Framework with Structured Weight Pruning and Quantization Using ADMM
    Yuan, Geng
    Ma, Xiaolong
    Ding, Caiwen
    Lin, Sheng
    Zhang, Tianyun
    Jalali, Zeinab S.
    Zhao, Yilong
    Jiang, Li
    Soundarajan, Sucheta
    Wang, Yanzhi
    2019 IEEE/ACM INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN (ISLPED), 2019,
  • [8] Fast and Accurate Output Error Estimation for Memristor-Based Deep Neural Networks
    Kern, Jonathan
    Henwood, Sebastien
    Mordido, Goncalo
    Dupraz, Elsa
    Aissa-El-Bey, Abdeldjalil
    Savaria, Yvon
    Leduc-Primeau, Francois
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2024, 72 : 1205 - 1218
  • [9] Alternative memristor-based interconnect topologies for fast adaptive synchronization of chaotic circuits
    Escudero, Manuel
    Vourkas, Ioannis
    Rubio, Antonio
    CHAOS SOLITONS & FRACTALS, 2020, 138
  • [10] Purely self-rectifying memristor-based passive crossbar array for artificial neural network accelerators
    Kanghyeok Jeon
    Jin Joo Ryu
    Seongil Im
    Hyun Kyu Seo
    Taeyong Eom
    Hyunsu Ju
    Min Kyu Yang
    Doo Seok Jeong
    Gun Hwan Kim
    Nature Communications, 15