Reinforcement Learning Circuit Based on Memristive Neural Network

Operant conditioning plays a pivotal role in the realm of biology and serves as a fundamental concept in the field of artificial intelligence reinforcement learning. While mem-ristive neural circuits related to operant conditioning have been explored, there is a dearth of research addressing the influence of various reinforcers and reinforcement methods on the learning process. This paper presents a novel operant conditioning neural network designed to learn using secondary reinforcers and spaced stimuli. It not only facilitates the acquisition of primary and secondary reinforcer learning but also investigates the disparate learning outcomes resulting from continuous reinforcement and interval reinforcement in operant conditioning. The proposed circuit's functionality is rigorously validated through PSPICE simulation, affirming its feasibility. This work provides an implementation idea for the hardware implementation of artificial intelligence.