期刊:Journal of Nanoelectronics and Optoelectronics [American Scientific Publishers] 日期:2022-03-01卷期号:17 (3): 505-515
标识
DOI:10.1166/jno.2022.3222
摘要
Associative learning and memory are widely studied, but non associative learning is rarely discussed. A non associative learning circuit based on memristor is proposed in this paper, including habituation, sensitization, presynaptic facilitation, secondary habituation, long and short term habituation, cross habituation, long and short timescale habituation. Habituation and sensitization in non associative learning are achieved through memristive non-volatility and non-linearity, and presynaptic facilitation is also discussed. Long and short term habituation and secondary habituation are considered through the synaptic module and the voltage module. Cross habituation and long and short timescale habituation are introduced through the suppression module and synaptic module. The more realistic brain is simulated by artificial neural network through the study of non associative learning.