Beyond Reinforcement Learning: Fast and Scalable Quantum Circuit Synthesis

TL;DR

This paper introduces a fast, scalable quantum circuit synthesis method that combines supervised learning with stochastic beam search, achieving high success rates and reduced training costs for complex quantum circuits.

Contribution

It presents a novel approach that uses supervised learning to estimate minimal description length and integrates it with stochastic beam search, enabling zero-shot generalization and faster synthesis.

Findings

Achieves faster synthesis times than previous methods.

Outperforms state-of-the-art in success rate for complex circuits.

Reduces training overhead with a lightweight model.

Abstract

Quantum unitary synthesis addresses the problem of translating abstract quantum algorithms into sequences of hardware-executable quantum gates. Solving this task exactly is infeasible in general due to the exponential growth of the underlying combinatorial search space. Existing approaches suffer from misaligned optimization objectives, substantial training costs and limited generalization across different qubit counts. We mitigate these limitations by using supervised learning to approximate the minimum description length of residual unitaries and combining this estimate with stochastic beam search to identify near optimal gate sequences. Our method relies on a lightweight model with zero-shot generalization, substantially reducing training overhead compared to prior baselines. Across multiple benchmarks, we achieve faster wall-clock synthesis times while exceeding state-of-the-art methods in terms of success rate for complex circuits.