AQC 2016 - A Fully-Programmable Measurement-Feedback OPO Ising Machine with All-to-All Connectivity

A Google TechTalk, June 28, 2016, presented by Peter McMahon (Stanford) ABSTRACT: An annealing machine for solving binary combinatorial optimization problems using networks of OPOs was demonstrated with 4 OPOs in 2014 [1]. We have recently extended this concept in two ways: firstly we have realized a network of 100 OPOs, and secondly we have used a measurement-feedback-based scheme to allow coupling of any OPO to any other OPO, allowing the realization of all-to-all couplings in an Ising Hamiltonian. In this presentation, we will outline the design of our experimental apparatus, and show results of trying to find ground states of various Ising problem instances with up to 100 spins and 10,000 connections. [1] A. Marandi, et al. doi:10.1038/nphoton.2014.249 Peter McMahon (Stanford University), Alireza Marandi (Stanford University), Yoshitaka Haribara (NII), Ryan Hamerly (Stanford University), MM Fejer (Stanford University), Hideo Mabuchi (Stanford University), Yoshihisa Yamamoto (Stanford University and JST) Presented at the Adiabatic Quantum Computing Conference, June 26-29, 2016, at Google's Los Angeles office.