Experiment shows how software-optimized circuits run less error-prone quantum algorithms

A research partnership at Lawrence Berkeley National Laboratory’s (Berkeley Lab) Advanced Quantum Testbed (AQT) and Chicago-based Super.tech (acquired by ColdQuanta in May 2022) demonstrated how to optimize the ZZ network protocol execution SWAP, important for quantum computing. The team also introduced a new quantum error reduction technique that will improve network protocol implementation in quantum processors. The experimental data were published in July i Physical examination researchh, adding more short-term ways to implement quantum algorithms using gate-based quantum computing.

A smart compiler for superconducting quantum hardware

Quantum processors with two- or three-dimensional architectures have limited qubit connectivity, where each qubit only interacts with a limited number of other qubits. In addition, the information in each qubit can only exist for so long before noise and errors cause decoherence, limiting the running time and reliability of quantum algorithms. Therefore, when designing and implementing a quantum circuit, researchers must optimize the translation of the circuit, which consists of abstract (logic) gates, into physical instructions based on the native hardware gates available in a given quantum processor. Efficient circuit decompositions minimize runtime because they account for the number of gates and operations natively supported by the hardware to perform the desired logic operations.

SWAP gates – which exchange information between qubits – are often introduced in quantum circuits to facilitate interactions between information in non-adjacent qubits. If a quantum device only allows gates between adjacent qubits, permutations are used to move information from one qubit to another non-adjacent qubit.

In noisy intermediate-scale quantum hardware (NISQ), the introduction of swap-gates can require significant experimental overhead. The exchange port often needs to be decomposed into native ports, such as NON-controlled ports. Therefore, when designing quantum circuits with limited qubit connectivity, it is important to use a smart compiler that can find, decompose, and cancel redundant quantum gates to improve the runtime of an algorithm or quantum application.

The research partnership used Super.tech’s SuperstaQ software, which allowed scientists to fine-tune their applications and automate circuit builds for AQT’s superconducting hardware, specifically for a built-in high-fidelity controlled S-gate, which is not available on most hardware systems. This clever compilation approach with four transmon qubits allows SWAP networks to be decomposed more efficiently than standard decomposition methods.

A ZZ SWAP gate array requires only minimal linear connection between qubits without additional couplings, thus providing practical advantages for efficient execution of quantum algorithms such as the quantum approximate optimization algorithm (QAOA). QAOA approximates solutions to combinatorial optimization problems – finding the optimal answer given a set of criteria. The Maximum-Cut problem, which can be used to arrange hubs on a transportation network system, is an example of a famous combinatorial optimization problem that can potentially be solved faster with QAOA using quantum circuits.

“One of the most difficult challenges in quantum computing is performing discrete logic operations. Because our control signals are analog and continuous, they are always imperfect. As we build more complex quantum circuits, the software framework that optimally compiles gates to match AQT’s hardware helps , us with achieving greater operational reliability,” Akel Hashim, principal investigator of ‘AQT on experience and graduate student at the University of California, Berkeley .

“A unique property of quantum computing is that it allows partial logic gates. This feature has no equivalent in traditional Boolean logic—for example, your laptop cannot perform 50% of an AND gate. AQT’s ability to calibrate these partially controlled S -quantum ports have allowed us to develop a wider range of new optimizations to get the most out of the hardware,” said Rich Rines, formerly of Super.tech and currently a software engineer at ColdQuanta.

“A key software engineering challenge for this experiment was remote collaboration, so we iteratively developed quantum circuit optimizations based on the custom gates calibrated by the AQT team. We optimized end-to-end by figuring out how to serialize these pulses while considering the hardware .We also figured out how to integrate open source quantum software packages into our compiler to ensure our optimizations don’t reinvent the wheel,” said Victory Omole, former Super.tech and software engineer at ColdQuanta.

As part of the experiment, the team also introduced a new technique called Equivalent Circuit Averaging (ECA), which randomized the various parameters of SWAP networks to generate many logically equivalent circuits. ECA randomizes the breakdown of quantum circuits, reducing the impact of systematic coherence errors – one of the most serious errors in quantum computing and widely studied at AQT.

“I proposed a way to merge my previous experimental work in random compilation with Quantum Benchmark (acquired by Keysight) using Super.tech’s Smart Compiler to investigate a new way to reduce the impact of crosstalk errors,” Hashim said. “I would not have had the insight to come up with this idea if I had not worked together with other researchers through the AQT user program. As someone about to enter the workforce, networking is key to building a core group of people I know in the field who are experts in various fields to whom I can also pitch research ideas. »

These experimental optimizations have improved the accuracy of the QAOA performance by up to 88%. Researchers seek to continue to explore and refine the methods of this work and apply them to other applications.

Support industry growth with an open access research laboratory

AQT operates a state-of-the-art open experimental test bed based on superconducting circuits and is funded by the US Department of Energy’s Advanced Scientific Computing Research (ASCR) program. Technologies developed elsewhere can be implemented and field tested at AQT, providing deep access to the full quantum computing stack at no additional cost.

Since launching its user program in 2020, AQT has given Super.tech, one of many industrial users, low-level access to hardware to test its ideas. Few cloud-based quantum platforms offer this kind of comprehensive access to the entire quantum computing stack and real-time feedback from hardware experts at no cost. Super.tech has collaborated with AQT’s experimental team of experts to learn how to improve performance on this type of hardware.

“By revealing the internal controls of quantum hardware, AQT’s collaborative approach with users is driving innovation across the quantum computing stack. We look forward to continuing our research collaboration with AQT, and we will continue to share these findings with the scientific community by publishing our experiences .” said Pranav Gokhale, vice president of Quantum Software at ColdQuanta and former CEO and co-founder of Super.tech.

Berkeley Lab’s AQT continues to grow as a cutting-edge center for quantum information research and development by bringing together expertise and users, including early-stage startups such as Super.tech, which now continues its journey of growth as part of ColdQuanta.

Leave a Comment