A novel approach for assessing quantum gate inaccuracies

Researchers at USC have formulated a new procedure for benchmarking quantum gates, a vital step towards unlocking the full capabilities of quantum computing and potentially accelerating advancements towards fault-tolerant quantum computers, as illustrated in a research paper recently published in Chemical Reviews.

The innovative protocol, termed deterministic benchmarking (DB), offers a more comprehensive and effective approach for pinpointing specific forms of quantum noise and inaccuracies compared to commonly utilized existing methods.

“Quantum computing is fundamentally constrained by how accurately we can execute gates — the essential operations of a quantum processor,” commented Daniel Lidar, co-corresponding author of the research and professor of electrical and computer engineering, chemistry, and physics and astronomy at the USC Viterbi School of Engineering and the USC Dornsife College of Letters, Arts and Sciences. “Our new method can detect both coherent and incoherent types of errors using merely a few straightforward experiments, making it significantly more efficient than current methodologies.”

Quantum gates and inaccuracies

Quantum computing holds the potential to solve intricate problems that exceed the capabilities of traditional, or classical, computers. Nevertheless, the precision of quantum calculations is heavily reliant on the functioning of quantum gates, which are vulnerable to inaccuracies stemming from noise and miscalibration.

Quantum gates carry out operations on qubits, which serve as the quantum analogs of classical computer bits; they are crucial for developing quantum algorithms and form the foundational elements of quantum circuits and computations. They empower quantum computers to execute algorithms that can be exponentially quicker than those undertaken by classical computers for certain operations.

However, quantum gates are prone to noise and inaccuracies, which is why benchmarking and error correction are fundamental research areas in quantum computing. The two primary categories are coherent and incoherent errors. Coherent errors are deterministic and repeatable errors that maintain quantum state purity. Coherent errors accumulate as amplitudes, as opposed to probabilities, potentially leading to quadratically faster error escalation than incoherent errors.

Incoherent errors are a category of inaccuracies that arise from the interaction of quantum systems with the environment; these errors strip quantum computers of their quantumness, causing them to perform no better than classical computing systems.

Recently, physicists have recognized the significant role coherent errors play in constraining the performance of quantum computers. Eli Levenson-Falk, co-corresponding author of the study and assistant professor of physics and astronomy, as well as electrical and computer engineering at USC Dornsife, highlights the importance of precise benchmarking of gate inaccuracies.

“What’s distinctive about our method is that it can effectively differentiate between various forms of quantum errors,” Levenson-Falk stated. “This is vital because certain types of errors, especially coherent ones, can be more harmful to quantum algorithms and necessitate distinct mitigation tactics.”

Deterministic benchmarking enhances efficiency

Quantum benchmarks are a collection of protocols and strategies used to assess the functionality of the overall quantum computer, which encompasses its gates, circuits, and processors. These protocols
“““html

are essential to the advancement and enhancement of quantum computing technologies by offering quantitative assessments of the performance of quantum operations amidst noise and inaccuracies.

Lidar remarked that deterministic benchmarking represents a notable progress in quantum computing due to its predictable and efficient nature. Unlike other benchmarking approaches, DB employs a small, fixed collection of straightforward pulse-pair sequences instead of averaging random circuits.

The researchers indicated that the crucial aspect of grasping the significance of DB is to juxtapose it with randomized benchmarking (RB), a commonly utilized technique for estimating the mean error rate of quantum gates. In contrast to RB, which averages numerous random gate sequences to yield a singular error measurement, DB utilizes crafted sequences to identify specific sources of error that may go undetected when employing RB.

Innovative method paves the way for progress in quantum chemistry and materials science

The researchers showcased DB on a superconducting transmon qubit — a prevalent type of superconducting qubit in quantum computing — to illustrate its capability to identify minor variations in qubit parameters that remain hidden to conventional benchmarking methods.

“By conducting multiple experiments, we showcased the diverse capabilities of DB,” Lidar stated. He noted that the exceptional feature of DB is its ability to deliver comprehensive insights into both coherent and incoherent errors, facilitating improved calibration of quantum gates. Furthermore, DB necessitates fewer experimental iterations than RB, thereby enhancing resource efficiency.

The findings hold substantial ramifications for applications in quantum chemistry and materials science, where accurate gate operations are vital for achieving dependable simulations of molecular systems.

The researchers intend to investigate avenues to extend DB to two-qubit gates, which could enable the creation of more sophisticated quantum circuits. Moreover, they are examining how DB can be tailored for other quantum computing architectures beyond superconducting qubits, such as trapped ions and photonic systems.

---

About the study: In addition to Levenson-Falk and Lidar, other contributors to the study include Vinay Tripathi and Daria Kowsari (co-lead authors), along with Kumar Saurav and Haimeng Zhang, all from USC.

This research received funding from the National Science Foundation, the Quantum Leap Big Idea Grant No. OMA-1936388, the Army Research Office MURI Grant No. W911NF-22-S-0007, and the Intelligence Advanced Research Projects Activity (IARPA) under Cooperative Agreement No. W911NF-23-2-0216.