Circuit Compilation Optimises Gate Count for Ion Trap Quantum Computers

Adapting quantum algorithms for real-world hardware demands efficient circuit compilation, a process that translates abstract instructions into operations a computer can actually perform, and Alejandro Villoria, Henning Basold, and Alfons Laarman from the Leiden Institute of Advanced Computer Science have developed a new approach to significantly improve this process. Their research focuses on optimising quantum circuits by incorporating ‘global gates’, operations uniquely available on certain hardware like ion trap computers, which allow multiple qubits to interact simultaneously. The team designed an algorithm that transforms any quantum circuit into an equivalent one utilising these global gates, effectively reducing the number of interactions needed and improving performance. Benchmarking against existing methods demonstrates that this new algorithm offers a substantial advantage when considering the limitations and capabilities of current quantum hardware.

Reducing the number of quantum gates required to execute a circuit represents one such optimisation strategy; methods for minimising non-Clifford gates or CNOT gates have attracted considerable research interest. For specific hardware platforms, such as ion trap quantum computers, it is possible to leverage unique properties to further reduce the cost of executing a quantum circuit. This work explores these possibilities, aiming to optimise circuit execution on such platforms

Global Mapping Simplifies Quantum Circuit Compilation

Researchers have developed a new compiler, named GMS Compiler, that optimises quantum circuits for hardware with limited connectivity and gate sets. The goal is to reduce the number of gates, particularly two-qubit gates, and rearrange the circuit for efficient execution on near-term quantum devices. This compiler focuses on leveraging global interactions to achieve better optimisation. The compiler utilizes ZX-Calculus, a graphical language for reasoning about quantum circuits, to transform and simplify circuits. The initial step involves converting the input circuit into a ZX-diagram.

A core component is Linear Programming, which finds optimal arrangements of gates, particularly CNOTs, within a portion of the circuit, aiming to minimize their number. The compiler works by iteratively processing a frontier of the circuit, optimizing it independently. This approach systematically optimizes circuits for improved performance on quantum hardware.

ZX-Calculus Optimizes Ion Trap Quantum Circuits

Researchers have developed a new method for compiling quantum circuits, specifically tailored for ion trap quantum computers, that significantly reduces the resources needed for computation. This approach leverages the unique capabilities of ion trap hardware, namely its ability to perform global interactions between qubits, to optimize circuit execution. Unlike traditional methods, this technique compiles circuits to utilize these global interactions, resulting in more efficient computations. The core of this advancement lies in the application of ZX-Calculus, a diagrammatic language for representing quantum computations, to design a compilation algorithm.

This algorithm extracts circuits and reshapes them to be more amenable to global operations, minimizing the number of these interactions required. By strategically grouping entangling gates, the algorithm reduces the overall complexity of the circuit and streamlines its execution on ion trap hardware. Benchmarking of the algorithm across various quantum circuits demonstrates substantial improvements over existing methods. The new approach effectively compiles arbitrary quantum circuits, a feat not achieved by many previous techniques limited to specific circuit types. This versatility is particularly important as it allows for broader application across different quantum algorithms and computational tasks. The results indicate a significant reduction in the number of global gates required, leading to more efficient and scalable quantum computations on ion trap systems.

This work presents a new algorithm for compiling quantum circuits specifically for ion trap quantum devices. The algorithm leverages the unique capabilities of this hardware, particularly global interactions like the Global Mølmer-Sørensen gate, to optimize circuit performance. By grouping entangling gates into these global operations, the researchers demonstrate improvements in circuit efficiency compared to both naive implementations and existing optimization tools. Future research could focus on modifying the circuit extraction process to further refine compilation algorithms and investigating the potential benefits of allowing different coupling strengths between qubits.