Emerging dynamic memristors for neuromorphic reservoir computing
114 Citations•2021•
Jie Cao, Xumeng Zhang, Hongfei Cheng
The critical characteristic parameters of memristors affecting the performance of RC systems, such as reservoir size and decay time, are identified and discussed, and the challenges this field faces in reliable and accurate task processing are summarized and forecasted.
Abstract
<jats:p>This work reviews the state-of-the-art physical reservoir computing systems based on dynamic memristors integrating with unique nonlinear dynamics and short-term memory behavior. The key characteristics, challenges and perspectives are also discussed.</jats:p>