Quantifying and Reducing the Overhead of Topological Quantum Error Correction in Large-Scale System

Опубликовано 28 июля 2016, 1:24

In this talk I will quantify the resources (number of qubits, number of quantum gates, computation time) needed to perform error correction with the surface code. Using examples of several models of quantum computation and quantum algorithms, I will demonstrate the advantages of the surface code compared to concatenated quantum error correcting codes. Although the surface code can tolerate very high error rates and uses fewer resources than concatenated codes, decoding errors requires solving a large instance of the minimum weight perfect matching problem. I will describe several simple heuristics that reduce the time complexity of error decoding to O(n), allowing us to decode errors in systems with up to million qubits on commodity hardware. I will also describe our recent progress on parallelizing the decoding algorithm.