Neurotransmitter‐Mediated Plasticity in 2D Perovskite Memristor for Reinforcement Learning

Abstract

The neuromorphic computing architecture is a promising artificial intelligence for implementing hierarchical processing, in‐memory computing, event‐driven operation and functional specialization in computing systems. However, current investigations mainly focus on unisensory processing without objective experience which is contrary to the flexible sensory learning capability in the human brain that can sense and process information according to the ever‐changing environment. For example, a dominant paradigm for reconfigurable bio‐learning features is the emotional experience. The neurotransmitter dopamine is released during arousal, influencing the vital brain functions involved in cognition, reward learning, movement and motivation. Here, the on‐demand configuration of a biorealistic synaptic connection based on a 2D CaTa2O7 (CTO) device is demonstrated that can be adaptively reconfigured for a reinforcement learning purpose by the light‐active resistive switching, which originated from the photon‐regulated metaplasticity. The low energy consumption of 12.4 fJ endows the reinforcement learning system with high power efficiency and reliability. Finally, in‐sensor computing with a CTO synapse is implemented with a filtering function to process digital data in a neuromorphic engineering manner. This work demonstrates the feasibility of 2D perovskite neuromorphic device with enhanced biological plausibility in the approaching post‐Moore era.

Element Regulation and Dimensional Engineering Co-Optimization of Perovskite Memristors for Synaptic Plasticity Applications

“…In addition, it is noteworthy that the quasi-2D BA 0.15 FA 0.85 PbI 3 perovskite occupies an intermediary position between pure 2D and 3D perovskites, combining the advantages of both structures. Furthermore, when compared to some kinds of materials including matel oxides, polymers, pure 2D perovskites, and pure 3D perovskites, the quasi-2D perovskite displays superior repeatability and enhanced stability. − In comparison to 3D perovskite configurations, the quasi-2D perovskite demonstrates a lower current flow and reduced power consumption. Compared with polymer and organic materials, the quasi-2D perovskites have a better endurance ability by a large margin.…”

“…The low energy consumption of 12.4 fJ ensures high power efficiency and in-sensor computing with a synapse showcasing the feasibility of 2D perovskite neuromorphic devices with enhanced biological plausibility in the post-Moore era. 26 Li et al introduce a hybrid 2D/3D perovskite heterostructure, achieved by depositing n-butylammonium iodide on the perovskite surface, enhancing resistive switching memory devices with improved performance, offering new possibilities for high-performance perovskitebased memory devices in next-generation technology. 27 Zhang et al explore 0D perovskite, 2D perovskite, and 1D perovskite films in memristors, demonstrating their exceptional stability, superior ON/OFF ratios, and dimensionality-dependent effects on resistive switching behaviors, while also simulating biological synapse plasticity and successfully emulating associative learning behavior.…”

“…Han et al demonstrate the adaptive, on-demand configuration of a biorealistic synaptic connection using a 2D perovskite device, exhibiting photon-regulated synaptic plasticity for reinforcement learning. The low energy consumption of 12.4 fJ ensures high power efficiency and in-sensor computing with a synapse showcasing the feasibility of 2D perovskite neuromorphic devices with enhanced biological plausibility in the post-Moore era . Li et al introduce a hybrid 2D/3D perovskite heterostructure, achieved by depositing n -butylammonium iodide on the perovskite surface, enhancing resistive switching memory devices with improved performance, offering new possibilities for high-performance perovskite-based memory devices in next-generation technology .…”

Recent Progress in Wearable Near-Sensor and In-Sensor Intelligent Perception Systems

“…Compared to the near-sensor visual system, the in-sensor visual system has been systematically researched due to the proposal of a wide variety of photo-induced resistive materials [147][148][149][150][151][152]. This means that neuromorphic devices can be modulated by an optical spike being directly dropped into the procedure of electric spike converting.…”

MoS2 Quantum Dot-Optimized Conductive Channels for a Conjugated Polymer-Based Synaptic Memristor

“…Memristors have the potential to overcome the limitations of Von Neumann systems in terms of their program operation modes, enabling them to efficiently process and store vast amounts of data. − Memristors have a simple sandwich structure similar to that of biological synapses. Biological research has shown that the human brain is made up of about 10 15 synapses, essentially making it a natural supercomputer that combines low-power consumption with high performance. − Brain-like biomimetic memristors are an important direction for future artificial intelligence computing applications, − and constructing memristor devices compatible with the current computing architectures and implementing neuromorphic computing is a promising approach.…”