Feynman-path type simulation using stabilizer projector decomposition of unitaries

TL;DR

This paper introduces a classical simulation approach for quantum circuits using stabilizer projector decomposition, bridging stabilizer methods and Feynman-path simulation, with potential advantages in specific parameter regimes.

Contribution

The paper presents two novel simulation algorithms, SPIR and SPC, based on stabilizer projector decomposition, and analyzes their performance compared to existing methods.

Findings

Identifies parameter regimes where the method outperforms others.

Provides estimates for simulating quantum supremacy experiments.

Discusses potential improvements for the proposed simulation techniques.

Abstract

We propose a classical simulation method for quantum circuits based on decomposing unitary gates into a sum of stabilizer projectors. By only decomposing the non-Clifford gates, we take advantage of the Gottesman-Knill theorem and build a bridge between stabilizer-based simulation and Feynman-path-type simulation. We give two variants of this method: stabilizer-based path-integral recursion (SPIR) and stabilizer projector contraction (SPC). We also analyze further advantages and disadvantages of our method compared to the Bravyi-Gosset algorithm and recursive Feynman path-integral algorithms. We construct a parametrized circuit ensemble and identify the parameter regime in this ensemble where our method offers superior performance. We also estimate the time cost for simulating quantum supremacy experiments with our method and motivate potential improvements of the method.