Pauli Network Circuit Synthesis with Reinforcement Learning
This work addresses the challenge of improving circuit quality for quantum transpilation workloads, offering a fast and effective optimization procedure with significant reductions in gate counts, though it is incremental as it builds on existing heuristic methods.
The paper tackles the problem of optimizing quantum circuits containing Pauli rotations and Clifford operations by introducing a reinforcement learning-based method for circuit re-synthesis, resulting in over 2x reduction in two-qubit gate count on random Pauli Networks and average improvements of 20% in gate count and depth across benchmarks.
We introduce a Reinforcement Learning (RL)-based method for re-synthesis of quantum circuits containing arbitrary Pauli rotations alongside Clifford operations. By collapsing each sub-block to a compact representation and then synthesizing it step-by-step through a learned heuristic, we obtain circuits that are both shorter and compliant with hardware connectivity constraints. We find that the method is fast enough and good enough to work as an optimization procedure: in direct comparisons on 6-qubit random Pauli Networks against state-of-the-art heuristic methods, our RL approach yields over 2x reduction in two-qubit gate count, while executing in under 10 milliseconds per circuit. We further integrate the method into a collect-and-re-synthesize pipeline, applied as a Qiskit transpiler pass, where we observe average improvements of 20% in two-qubit gate count and depth, reaching up to 60% for many instances, across the Benchpress benchmark. These results highlight the potential of RL-driven synthesis to significantly improve circuit quality in realistic, large-scale quantum transpilation workloads.