AQC 2016 - Roadmap for Building a Quantum Computer

A Google TechTalk, June 28, 2016, presented by John Martinis (Google) ABSTRACT: I will overview the basic strategy and roadmap for the quantum-AI project at Google, which has the goal of building a useful quantum computer. For hardware, the key metric is building scalable qubits with 2-qubit gate errors below 0.1-0.2% [J.M.Martinis, NPJQI 1, 15005 (2015)]. For software, I will describe a new "quantum-supremacy" test that can demonstrate the exponential power of a quantum processor by checking its output with a classical computer, which is intractable for even the world's most advanced classical supercomputer beyond 42-50 qubits. We are working to perform this experiment in the next 2 years. Presented at the Adiabatic Quantum Computing Conference, June 26-29, 2016, at Google's Los Angeles office.