Demonstration of Neuromodulation‐inspired Stashing System for Energy‐efficient Learning of Spiking Neural Network using a Self‐Rectifying Memristor Array. (31st March 2022)

• Record Type: Journal Article
• Authors: Cheong, Woon Hyung; Jeon, Jae Bum; In, Jae Hyun; Kim, Geunyoung; Song, Hanchan; An, Janho; Park, Juseong; Kim, Young Seok; Hwang, Cheol Seong; Kim, Kyung Min
• Is Part Of: Advanced functional materials. Volume 32:Number 29(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-03-31
• Subjects: memristors -- neural network -- neuromodulation -- neuromorphic devices -- stashing system; Materials -- Periodicals; Chemical vapor deposition -- Periodicals; 620.11
• DOI: 10.1002/adfm.202200337
• Languages: English
• ISSNs: 1616-301X

Abstract

Abstract: Neuromorphic engineering aims to mimic brain functions to achieve energy‐efficient artificial intelligence. Since researchers have indicated that memristors can mimic synapses and neurons, various studies have demonstrated the operation of neural networks using memristive dot product engine (MDPE) hardware. However, although several feasible implementations of synapse and neuron behaviors have been reported, few studies have demonstrated the system‐level energy‐efficient operation on the hardware. This work proposes a novel system inspired by the neuromodulation of the brain, referred to as a "stashing system." In the system, the trained synapses are stashed temporarily during the training of the spiking neural network and then merged for inferencing. The software simulation first confirmed the working principle of the stashing system. Then, a hardware demonstration is performed at an integrated 32 × 32 MDPE embodying a self‐rectifying and electroforming‐free memristor cell to validate the system. The results confirm that energy consumption in the memristor array is reduced by 37% for the unsupervised learning of the MNIST dataset. Abstract : A unified hardware and system solution for energy‐efficient and lightweight neuromorphic computing is proposed. A selector‐less 32 × 32 memristive crossbar array embodying a charge trap memristor with a four‐bits‐per‐cell operation is demonstrated. A novel algorithm for the stashing system inspired by neuromodulation in the brain is developed. The method can reduce energy consumption by ≈37%.