Information Dimension Matching in Memristive Computing System for Analog Deployment of Deep Neural Networks
Authors:
Open Access
- 28 May 2024
- journal article
- research article
- Published by Wiley in Advanced Electronic Materials
- Vol. 10 (10)
Abstract
Memristor, with the ability of analog computing, is widely investigated for improving the computing efficiency of deep neural networks (DNNs) deployment. However, how to fully take advantage of the analog computing ability of memristive computing system (MCS) for DNN deployment is still an open question. Here, a new neural network models deployment scheme, that is, an information dimension matching (IDM) scheme, is proposed to fully take advantage of the analog computing ability of MCS. Furthermore, the spatial and temporal DNN, that is convolutional neural network (CNN) and recurrent neural network (RNN) is used to verify the proposed deployment scheme, respectively. The experimental results indicate that, compared to the traditional deployment schemes, the proposed deployment scheme shows obvious inference accuracy and energy efficiency improvement (>4 × in four‐layer DNNs deployment), and the energy efficiency improvement increases dramatically with the layers increment of DNNs. This work paves the path for developing high computing efficiency analog MCS.Keywords
Funding Information
- National Natural Science Foundation of China (62304001, 62274002, 61874001, 62004001, 62201005)
- Natural Science Foundation of Anhui Province (2308085QF213)
This publication has 27 references indexed in Scilit:
- Implementing in-situ self-organizing maps with memristor crossbar arrays for data mining and optimizationNature Communications, 2022
- Device Variation Effects on Neural Network Inference Accuracy in Analog In‐Memory Computing SystemsAdvanced Intelligent Systems, 2022
- 4K-memristor analog-grade passive crossbar circuitNature Communications, 2021
- Hybrid memristor-CMOS neurons for in-situ learning in fully hardware memristive spiking neural networksScience Bulletin, 2021
- Memristor crossbar architectures for implementing deep neural networksComplex & Intelligent Systems, 2021
- A Habituation Sensory Nervous System with MemristorsAdvanced Materials, 2020
- Memory devices and applications for in-memory computingNature Nanotechnology, 2020
- Efficient and self-adaptive in-situ learning in multilayer memristor neural networksNature Communications, 2018
- Equivalent-accuracy accelerated neural-network training using analogue memoryNature, 2018
- Training and operation of an integrated neuromorphic network based on metal-oxide memristorsNature, 2015