A Learning‐Rate Modulable and Reliable TiOx Memristor Array for Robust, Fast, and Accurate Neuromorphic Computing. Issue 22 (5th June 2022)

• Record Type: Journal Article
• Authors: Jang, Jingon; Gi, Sanggyun; Yeo, Injune; Choi, Sanghyeon; Jang, Seonghoon; Ham, Seonggil; Lee, Byunggeun; Wang, Gunuk
• Is Part Of: Advanced science. Volume 9:Issue 22(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-06-05
• Subjects: artificial synapses -- hardware implementation -- memristors -- neuromorphic computing -- uniformity; Science -- Periodicals; 505
• DOI: 10.1002/advs.202201117
• Languages: English
• ISSNs: 2198-3844

Abstract

Abstract: Realization of memristor‐based neuromorphic hardware system is important to achieve energy efficient bigdata processing and artificial intelligence in integrated device system‐level. In this sense, uniform and reliable titanium oxide (TiO x ) memristor array devices are fabricated to be utilized as constituent device element in hardware neural network, representing passive matrix array structure enabling vector‐matrix multiplication process between multisignal and trained synaptic weight. In particular, in situ convolutional neural network hardware system is designed and implemented using a multiple 25 × 25 TiO x memristor arrays and the memristor device parameters are developed to bring global constant voltage programming scheme for entire cells in crossbar array without any voltage tuning peripheral circuit such as transistor. Moreover, the learning rate modulation during in situ hardware training process is successfully achieved due to superior TiO x memristor performance such as threshold uniformity (≈2.7%), device yield (> 99%), repetitive stability (≈3000 spikes), low asymmetry value of ≈1.43, ambient stability (6 months), and nonlinear pulse response. The learning rate modulable fast‐converging in situ training based on direct memristor operation shows five times less training iterations and reduces training energy compared to the conventional hardware in situ training at ≈95.2% of classification accuracy. Abstract : The uniform and reliable 25 × 25 TiOx memristor array device shows superior performance as threshold uniformity (≈2.7 %), device yield (> 99%), repetitive stability (≈3000 spikes), low asymmetry (≈1.43), ambient stability (6 months) and nonlinear pulse response. The integrated neuromorphic hardware system is implemented with 5 times less training cost than conventional system at ≈95.2% of accuracy.