Compact Modeling and Mitigation of Parasitics in Crosspoint Accelerators of Neural Networks
- 6 February 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71 (3) , 1900-1906
Abstract
In-memory computing (IMC) can accelerate data-intensive tasks, such as matrix-vector multiplication (MVM) or artificial neural networks (ANNs) inference, by means of the crosspoint memory array, allowing to reduce time and energy consumption. IMC accuracy, however, is affected by nonidealities, such as variability of the conductive weights or IR drop along wires due to parasitic resistances, whose impact steeply increases with the increase of array size. This work proposes a compact model to assess the impact of nonidealities for various circuital implementations, together with architectural schemes for their mitigation based on replicated arrays. The proposed mitigation techniques allow to restore the ANN accuracy from 72.7% to 94.9%, close to the software accuracy of 96.9%, in view of an increased area and energy consumption.Keywords
Funding Information
- ECSEL Joint Undertaking (101007321)
- European Union’s Horizon 2020 Research and Innovation Program through JU and France, Belgium, Czech Republic, Germany, Italy, Sweden, Switzerland, and Turkey
This publication has 21 references indexed in Scilit:
- Mitigate IR-Drop Effect by Modulating Neuron Activation Functions for Implementing Neural Networks on Memristor Crossbar ArraysIEEE Electron Device Letters, 2023
- Nonideality‐Aware Training for Accurate and Robust Low‐Power Memristive Neural NetworksAdvanced Science, 2022
- Parasitic-Aware Modeling and Neural Network Training Scheme for Energy-Efficient Processing-in-Memory With Resistive Crossbar ArrayIEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2022
- Effect of Program Error in Memristive Neural Network With Weight QuantizationIEEE Transactions on Electron Devices, 2022
- Modeling and Compensation of IR Drop in Crosspoint Accelerators of Neural NetworksIEEE Transactions on Electron Devices, 2022
- Redundancy and Analog Slicing for Precise In-Memory Machine Learning—Part I: Programming TechniquesIEEE Transactions on Electron Devices, 2021
- Accurate Program/Verify Schemes of Resistive Switching Memory (RRAM) for In-Memory Neural Network CircuitsIEEE Transactions on Electron Devices, 2021
- Analysis of the statistics of device-to-device and cycle-to-cycle variability in TiN/Ti/Al:HfO2/TiN RRAMsMicroelectronic Engineering, 2019
- Compressed Sensing With Approximate Message Passing Using In-Memory ComputingIEEE Transactions on Electron Devices, 2018
- In-memory computing with resistive switching devicesNature Electronics, 2018