Photonics for artificial intelligence and neuromorphic computing
Recent advances in integrated photonic neuromorphic neuromorphic systems are reviewed, current and future challenges are discussed, and the advances in science and technology needed to meet those challenges are outlined.
Abstract
Research in photonic computing has flourished due to the proliferation of\noptoelectronic components on photonic integration platforms. Photonic\nintegrated circuits have enabled ultrafast artificial neural networks,\nproviding a framework for a new class of information processing machines.\nAlgorithms running on such hardware have the potential to address the growing\ndemand for machine learning and artificial intelligence, in areas such as\nmedical diagnosis, telecommunications, and high-performance and scientific\ncomputing. In parallel, the development of neuromorphic electronics has\nhighlighted challenges in that domain, in particular, related to processor\nlatency. Neuromorphic photonics offers sub-nanosecond latencies, providing a\ncomplementary opportunity to extend the domain of artificial intelligence.\nHere, we review recent advances in integrated photonic neuromorphic systems,\ndiscuss current and future challenges, and outline the advances in science and\ntechnology needed to meet those challenges.\n