[16a-A410-1~8] FS.1 フォーカストセッション「AIエレクトロニクス」

▲ [16a-A410-8] Boosting learning ability of overdamped bistable stochastic resonance system based physical reservoir computing model by time-delayed feedback

〇(M2)Zhuozheng Shi¹、Zhiqiang Liao¹、Hitoshi Tabata¹ (1.Univ. of Tokyo)

キーワード：Stochastic Resonance, Reservoir Computing, Overdamped Bistable System

A time-delayed feedback overdamped bistable RC system (TOBRC) is proposed in this work, which is implemented in a physical environment with time-division nodes and simulated on a conventional computer. Short-term memory and parity testing tasks were conducted to evaluate the short-term memory capability and nonlinearity, which are the two core capabilities of physical RC for learning