The rapid advancement of artificial intelligence technology has pushed the limits of computing systems capable of managing large-scale data. However, conventional von Neumann–based architectures are inadequate for such workloads because their physically separated processing and memory units create bottlenecks in speed and energy efficiency. An alternative and promising method for massively parallel, low-power information processing is neuromorphic computing, inspired by the human brain’s function. Artificial synapses primarily enable information processing and transmission, traditionally through electrical signaling. Optical signaling, however, provides key performance advantages, including minimal electrical crosstalk, high operational speed, and reduced power consumption. These properties make optoelectronic synapses preferable to purely electronic ones for high-speed and energy-efficient computing. Semiconductor materials such as transition-metal dichalcogenides, metal oxides, and organic materials, including halide perovskites, have been explored for artificial synapses. Specifically, halide perovskite provides large optical cross-sections, mixed ionic–electronic conduction, tunable band gaps, and inherent current-voltage hysteresis, positioning them as strong candidates for optoelectronic synapses.

The symposium invites contributions on recent progress and challenges in materials design and device structures for low-power optoelectronic neuromorphic devices; structure-property relationships; mechanisms of artificial synapses and their applications; and emerging fields of quantum neuromorphic photonics.