RIA: Parallel Optical Architectures and Algorithms for Symbolic Artificial Intelligence Computing

本研究探索光学系统在符号人工智能计算中的应用。它同时涉及算法和体系结构。符号AI计算的基本基本运算和高度并行的新型光学体系结构正在开发中。使用分析建模方法、模拟技术和实验室实验相结合的方法来应用和分析启发式搜索程序。符号AI计算对速度和性能的要求越来越高，这对计算机体系结构和算法的设计提出了更高的要求。光计算是一种非常有前途的技术，可以将处理速度提高到电子技术所能提供的几个数量级。这项研究将导致新的光学体系结构和算法的发展，表现出高速和并行性。它还将使人们更好地理解光学在计算中的作用。

This research explores the application of optical systems to symbolic artificial intelligence computing. It addresses both algorithms and architectures. Basic fundamental operations for symbolic AI computing and highly parallel and novel optical architectures are being developed. Heuristic search procedures are applied and analyzed using a combination of analytical modeling methods, simulation techniques, and laboratory experiments. The increasing demands of speed and performance in symbolic AI computing are imposing stringent demands on the design of computer architectures and algorithms. Optical computing is a very promising technology for elevating the processing speed to several orders of magnitude over what electronic technologies can offer. This research will result in the development of new optical architectures and algorithms that exhibit high speed and parallelism. It will also result in a better understanding of the role of optics in computing.