The last lecture course that Nobel Prize winner Richard P. Feynman gaveto students at Caltech from 1983 to 1986 was not on physics but on computerscience. The first edition of the Feynman Lectures on Computation, publishedin 1996, provided an overview of standard and not-so-standard topics incomputer science given in Feynman’s inimitable style. Although nowover 20 years old, most of the material is still relevant and interesting, andFeynman’s unique philosophy of learning and discovery shines through.For this new edition, Tony Hey has updated the lectures with an invitedchapter from Professor John Preskill on “Quantum Computing 40 YearsLater”. This contribution captures the progress made toward building aquantum computer since Feynman’s original suggestions in 1981. The last25 years have also seen the “Moore’s law” roadmap for the IT industrycoming to an end. To reflect this transition, John Shalf, Senior Scientistat Lawrence Berkeley National Laboratory, has contributed a chapteron “The Future of Computing beyond Moore’s Law”. The final updatefor this edition is an attempt to capture Feynman’s interest in artificialintelligence and artificial neural networks. Eric Mjolsness, now a Professorof Computer Science at the University of California Irvine, was a TeachingAssistant for Feynman’s original lecture course and his research interestsare now the application of artificial intelligence and machine learningfor multi-scale science. He has contributed a chapter called “Feynmanon Artificial Intelligence and Machine Learning” that captures the earlydiscussions with Feynman and also looks toward future developments.This exciting and important work provides key reading for students andscholars in the fields of computer science and computational physics.
