SHF: FET: Medium: Designing and Synthesizing a Quantum Circuit Compiler

点击翻译按钮获取中文摘要

Quantum computers can enable advances in quantum physics, chemistry, material design, optimization, machine learning, and beyond. Today, scientists and engineers around the world are racing to build larger, more-reliable quantum computers and demonstrate their effectiveness at evaluating increasingly complex quantum programs (circuits). Unfortunately, although computationally powerful, quantum computers are fundamentally noisy, and mitigating noise is the primary hurdle in unlocking the true potential of quantum computers. This project focuses on the circuit compiler, which optimizes a quantum circuit and maps the circuit to a physical quantum computer. Given the small size of current and future quantum computers and the noisiness of their computation, careful compilation is essential to produce accurate and reliable results on quantum computers. This project's novelties are twofold- the use of synthesis frameworks to automatically generate parts of the quantum compiler and adaptive runtime that tailor quantum circuits to enable higher reliability. The project's impacts are the development of compiler tools that a broad community of researchers can leverage to program, benchmark, and optimize quantum computing systems.Currently, the state of quantum hardware is in flux, there are many physical realizations of quantum computers, and physicists constantly experiment with novel devices and architectures. For every new physical substrate and every modification of a computer, one needs to modify or rewrite major compiler pieces to run experiments. This project aims to create a framework to automatically synthesize a quantum circuit optimizer, mapper, and runtime system as per the hardware constraints. It focuses on three thrusts: (1) automatic discovery and verification of circuit transformations to reduce circuit size; (2) synthesis of circuit mappers tailored to hardware-specific constraints and noise characteristics to increase the fidelity of quantum circuits; and (3) an efficient runtime framework that adapts the quantum circuits to the changes in hardware characteristics. This project develops novel circuit optimization methods by leveraging a rich set of tools designed for conventional program synthesis and optimization problems.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.