TY - GEN
T1 - Validation of spiking neural networks using resistive-switching synaptic device with spike-rate-dependent plasticity
AU - Bang, Suhyun
AU - Oh, Min Hye
AU - Kim, Min Hwi
AU - Kim, Tae Hyeon
AU - Lee, Dong Keun
AU - Choi, Yeon Joon
AU - Kim, Chae Soo
AU - Hong, Kyungho
AU - Cho, Seongjae
AU - Kim, Sungjun
AU - Park, Byung Gook
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/1
Y1 - 2020/1
N2 - In this work, we have developed a spiking neural network (SNN) using gradual resistive-switching random-access memory (RRAM) synaptic device. The fabricated RRAM devices demonstrated the characteristics of gradually changing conductance with voltage pulses under both positive and negative polarities, which is suitable for imitating the potentiation and depression functions of a biological synapse by an electron device. Featuring the gradual switching characteristics, spike-rate-dependent plasticity (SRDP) inspired by Bienenstock, Cooper, and Munro (BCM) learning rule was confirmed and modeled for synaptic modification in the SNN. Then, the supervised learning of MNIST patterns was performed on the simulated SNNs, by which it has been validated that the proposed resistive-switching synaptic device and SRDP synaptic modification rule can adjust weights accurately in cooperation without necessitating the conventional calculation-based learning scheme in the artificial neural networks (ANNs), such as error backpropagation.
AB - In this work, we have developed a spiking neural network (SNN) using gradual resistive-switching random-access memory (RRAM) synaptic device. The fabricated RRAM devices demonstrated the characteristics of gradually changing conductance with voltage pulses under both positive and negative polarities, which is suitable for imitating the potentiation and depression functions of a biological synapse by an electron device. Featuring the gradual switching characteristics, spike-rate-dependent plasticity (SRDP) inspired by Bienenstock, Cooper, and Munro (BCM) learning rule was confirmed and modeled for synaptic modification in the SNN. Then, the supervised learning of MNIST patterns was performed on the simulated SNNs, by which it has been validated that the proposed resistive-switching synaptic device and SRDP synaptic modification rule can adjust weights accurately in cooperation without necessitating the conventional calculation-based learning scheme in the artificial neural networks (ANNs), such as error backpropagation.
KW - Resistive-switching random-access memory
KW - Spike-rate-dependent plasticity
KW - Spiking neural network
KW - Synaptic device
UR - https://www.scopus.com/pages/publications/85083508567
U2 - 10.1109/ICEIC49074.2020.9051210
DO - 10.1109/ICEIC49074.2020.9051210
M3 - Conference contribution
AN - SCOPUS:85083508567
T3 - 2020 International Conference on Electronics, Information, and Communication, ICEIC 2020
BT - 2020 International Conference on Electronics, Information, and Communication, ICEIC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 International Conference on Electronics, Information, and Communication, ICEIC 2020
Y2 - 19 January 2020 through 22 January 2020
ER -