QSymb

Synthesis of Compact and Expressive Quantum-Circuit Optimizations.

A state-of-the-art rule synthesizer and optimizer for quantum circuit optimization. Given a gate set, QSymb automatically synthesizes both concrete rules and symbolic rules, and QSymb-Optimizer uses them as input to perform optimization. On the IBM-Eagle gate set, Qsymb strictly outperforms state-of-the-art rewrite-based optimizers (Qiskit, Guoq-Rewrite, Quartz, TKET, and Queso) in two-qubit-gate reduction on 90%, 67%, 82%, 85%, and 83% of standard quantum algorithm benchmarks, respectively; on Nam gate set, the corresponding rates are 88%, 74%, 81%, 86%, and 82.9%. It achieves final average two-qubit-gate reductions of 27.44% and 29.95%.

Key ideas

• Symbolic rewrite rules — a symbolic gate stands for infinitely many subcircuits; a symbolic-matrix constraint characterizes exactly when two fragments are equivalent.
• Canonical symbolic rules (L; S = S; R) — a compact generative core from which general symbolic rules are derived.
• Rule anchoring — turns the canonical core into optimization-effective rules.
• Guarantees — soundness (validation), non-derivability, and bounded completeness.

QSymb synthesizes (1) a small, non-derivable concrete rule set complete up to chosen size/qubit bounds, and (2) a small, expressive canonical symbolic rule set.

Layout

Path	Contents
SymbolicOptimizer/	rule synthesis + optimizer (Java)
qsymb_benchmarks/, benchmark.txt	benchmark circuits and the 135-circuit suite
*.sh	build, synthesis, optimization, ablation scripts
qsymb_plot_tools/, paper_results/	figure generation and precomputed results

Prerequisites

Native toolchain (for building outside the image):

• JDK 17 (built/run with --enable-preview)
• Maven >= 3.6 (Java libraries — antlr4, jgrapht, guava, gson, commons-*, opencsv, lombok — are fetched automatically)
• egglog-experimental 1.0.0 (binary on PATH)
• Python 3.10 with the packages in requirements.txt:

pip install -r requirements.txt

Reproducing the paper

See Artifact_README.md for full build and experiment instructions (rule synthesis & cost — Tables 1–3; optimization performance — Figs. 11–12; ablations — Figs. 13–14).