CCF: FET: Small: Toward Scalable Quantum Circuit Optimization with Structural Gadgets

Quantum computers hold promise for solving scientific challenges that remain fundamentally intractable for today’s most powerful classical computers. Realizing that promise, however, requires more than better hardware; it also requires software that enables scientists, engineers, and other domain experts to express their ideas and translate them into reliable and efficient quantum instructions that can run on real machines. This project helps turn theoretical quantum algorithms into practical quantum advantage by making quantum computing easier to use, more effective in practice, and accessible to a broader community of innovators beyond those with physics or computer science background. This matters to the U.S. national interests — it solidifies American leadership in a critical technology sector, catalyzes breakthroughs in chemistry and cryptography, and cultivates a highly skilled, multidisciplinary quantum workforce. Technically, this project aims to build a scalable, modular, open-source quantum compiler that performs co-optimization across logical circuits, physical circuits, hardware mapping, resource management, and error mitigation. This project centers on four structured circuit forms — phase polynomials, phase gadgets, symbolic Pauli gates, and linear reversible circuits — and studies how to optimize them both individually and together across the whole programs. The approaches include developing new intermediate representations that capture circuit structure beyond simple gate-by-gate rewriting, integrating optimization with routing and device constraints for superconducting, trapped-ion, and neutral-atom hardware, and applying AI-assisted methods such as reinforcement learning, diffusion-style generative models, and heuristic search to guide synthesis, pass selection, and reliability-aware optimization. The intended outcome is a cross-stack compiler infrastructure that reduces gate count and depth, lowers compilation overhead, and improves the fidelity and performance of compiled quantum programs. More importantly, this work moves beyond "black-box" optimization. By identifying why specific compilation techniques work synergistically, the project provides a foundational understanding of quantum program structure. This insight reduces the reliance on computationally expensive exhaustive searches and supports the development of more intuitive, effective heuristics for the broader quantum computing community. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2529338 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Zheng Zhang | Institution: Rutgers University New Brunswick, NEW BRUNSWICK, NJ | Award Amount: $599,878 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2529338 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2529338.html