Compiler Optimization for Quantum Computing Using Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning framework for quantum circuit compilation that combines techniques from different compilers, significantly improving expected fidelity in most cases compared to individual tools.

Contribution

It proposes a novel reinforcement learning-based approach to optimize quantum circuit compilation flows by integrating multiple compiler techniques within a flexible framework.

Findings

Outperforms individual compilers in 73% of cases regarding expected fidelity

Supports combination of techniques from IBM's Qiskit and Quantinuum's TKET

Framework is publicly available on GitHub

Abstract

Any quantum computing application, once encoded as a quantum circuit, must be compiled before being executable on a quantum computer. Similar to classical compilation, quantum compilation is a sequential process with many compilation steps and numerous possible optimization passes. Despite the similarities, the development of compilers for quantum computing is still in its infancy -- lacking mutual consolidation on the best sequence of passes, compatibility, adaptability, and flexibility. In this work, we take advantage of decades of classical compiler optimization and propose a reinforcement learning framework for developing optimized quantum circuit compilation flows. Through distinct constraints and a unifying interface, the framework supports the combination of techniques from different compilers and optimization tools in a single compilation flow. Experimental evaluations show that the proposed framework -- set up with a selection of compilation passes from IBM's Qiskit and Quantinuum's TKET -- significantly outperforms both individual compilers in 73% of cases regarding the expected fidelity. The framework is available on GitHub (https://github.com/cda-tum/MQTPredictor) as part of the Munich Quantum Toolkit (MQT).