REVIEW 10 cited by
Practical and efficient quantum circuit synthesis and transpiling with Reinforcement Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This paper demonstrates the integration of Reinforcement Learning (RL) into quantum transpiling workflows, significantly enhancing the synthesis and routing of quantum circuits. By employing RL, we achieve near-optimal synthesis of Linear Function, Clifford, and Permutation circuits, up to 9, 11 and 65 qubits respectively, while being compatible with native device instruction sets and connectivity constraints, and orders of magnitude faster than optimization methods such as SAT solvers. We also achieve significant reductions in two-qubit gate depth and count for circuit routing up to 133 qubits with respect to other routing heuristics such as SABRE. We find the method to be efficient enough to be useful in practice in typical quantum transpiling pipelines. Our results set the stage for further AI-powered enhancements of quantum computing workflows.
Forward citations
Cited by 10 Pith papers
-
Transpiler Autotuning with Predictive Models for Quantum Circuit Optimization
A learning-to-rank model over feature-model-sampled Qiskit transpiler pass configurations reliably outperforms Qiskit's fixed optimization levels on two-qubit gate reduction.
-
Shielded RL for Route-Charged Parity-Term Ordering in QEDA Phase Components
Shielded RL reordering of commuting phase terms cuts routed CNOT counts by 5.7–12.2% over search baselines on parity-walk QEDA components, but the proxy does not transfer to extraction-heavy or token/permutation circuits.
-
RubriQ: Rubric-Guided Group Relative Policy Optimization for Constraint-Aware Quantum Circuit Synthesis
A rubric-guided GRPO pipeline fine-tunes a 7B LLM to synthesize quantum circuits achieving 3.31x T-gate compression with <1% hardware-constraint violations, validated on IBM and IonQ processors.
-
Aligning Quantum Operators with Large Language Models
An LLM that reads a quantum operator as image-like patches can synthesize 4-qubit Pauli-rotation circuits at high success and obey English gate constraints.
-
Quantum circuits as a game: A reinforcement learning agent for quantum compilation and its application to reconfigurable neutral atom arrays
A transformer-based reinforcement learning agent learns to reconfigure atoms in neutral atom arrays and reduces estimated logarithmic infidelity by up to about 20% on benchmark circuits, including unseen ones.
-
Improved Quantum Computation using Operator Backpropagation
By classically backpropagating an observable through part of a quantum circuit, the authors reduce the quantum circuit depth and achieve lower error for expectation values in a 127-qubit XY-model simulation.
-
Spectral surgery and high-fidelity quantum state transfer in $XX$ chains
Spectral surgery on a uniform XX chain yields analytic spin chains that interpolate between uniform and Krawtchouk chains and achieve good-fidelity state transfer with bounded couplings.
-
Unitary Synthesis with AlphaZero via Dynamic Circuits
An AlphaZero-like RL agent can synthesize exact Clifford+T circuits for up to three qubits with ancilla, recovering known optimal decompositions and a 4-T Toffoli implementation.
-
Improving Figures of Merit for Quantum Circuit Compilation
A random forest model trained on real QPU execution results predicts quantum circuit execution quality with a Pearson correlation of 0.88 to 0.94, beating gate count, depth, expected fidelity, and ESP by an average of 49%.
-
Gate teleportation-assisted routing for quantum algorithms
The Routing with Teleported Gates (RTG) method selects teleportation paths through unused qubits to lower circuit depth by up to about 25% in selected benchmarks, though implemented reductions are smaller and depend o...
Discussion (0). Continue with ORCID to comment.