High Performance Compilation and Decoding of Surface Code Protected Circuits

ABSTRACT:
Quantum benchmarking has the goal to determine what size, quality, and configuration of quantum computers, if any, will enable the revolutionary capabilities of quantum algorithms. of quantum computing. The surface code, in its by now classical form, is the leading candidate for implementing large scale error-corrected quantum computations. Benchmarking surface code computers requires, to start with, fast and efficient compilation and decoding methods and software. First, in this talk, we take advantage of the regular layout common in quantum circuits to express them as 3D objects composed of tiles, where each tile is a 3D nearest-neighbor implementation of the quantum Toffoli gate. This scheme is applied prior to compilation and can be used as an intermediate step for both NISQ and error corrected circuits which use the surface code. Second, we will present a state of the art, high performance compiler for surface code quantum error correction. Our compiler translates any quantum circuit to a fault-tolerant one error corrected one that follows lattice surgery instructions. Finally, we will discuss ongoing work on the implementation and evaluation of surface code decoders using neural networks and belief propagation with stochastic circuit elements.