Machine learning-assisted design of high-performance perovskite photodetectors: a review
Xiaohui Li, Yongxiang Mai, Chunfeng Lan, Fu Yang, Putao Zhang, Shengjun Li
Abstract
Photodetectors (PDs) based on perovskite materials have become a strong contender for next-generation optical sensing. Because it has the advantages of high photoelectric conversion efficiency, broad spectral response, low cost, and easy preparation, it has a promising application in the field of optoelectronics. Machine learning (ML) is a branch of artificial intelligence that enables computer systems to improve performance from data through algorithms and statistical models automatically. Recently, it has been used in performance prediction and material screening of optoelectronic devices. As a result, combining ML and perovskite PDs has received much attention to optimize manufacturing processes and reduce processing costs. In this review, we provide a comprehensive review of recent research advances in the use of ML for perovskite devices, analyze the application of different types of perovskite materials in PDs, and discuss the feasibility and challenges of applying ML in perovskite PDs. This review outlines a visionary perspective and a roadmap for the progression of perovskite PDs towards unparalleled performance benchmarks, offering insights into the future trajectory of this promising technology.