A tunable coupler can toggle the qubit-qubit interactions on and off. The unwanted or unneeded (ZZ) interactions between two qubits are removed using higher energy levels in the coupler. Credit: Krantz Nanoart
MIT researchers have discovered a method to reduce the errors of two-qubit gates dramatically. It is an important improvement towards fully realizing quantum computation.
MIT Researchers have created significant progress on the path to quantum computation’s fullest realization by demonstrating a method that eliminates the common errors in the primary quantum algorithms’ operation, such as the two-qubit operation, also known as “gate.”
“Despite great progress toward to be able to conduct computations with low mistake charges with superconducting quantum parts (qubits), mistakes in two-qubit gates, one of the foundations of quantum computation, persist,” states Youngkyu Sung, an MIT graduate student in computer science and electrical engineering who is the principal author of a study that discusses this issue, which was that was published on June 16, 2021, in Physical Review X. “We have demonstrated a way to reduce those errors sharply.”
Quantum computers processing information is a delicate operation performed by fragile qubits that are vulnerable to decoherence, losing their quantum mechanical properties. In the previous work carried out by Sung and the research team with which he collaborates, MIT Engineering Quantum Systems Tunable couplers were suggested that allow researchers to switch two-qubit interaction on and off to regulate their operation without compromising the qubits’ fragility. The concept of tunable couplers represented significant advancements. It was mentioned as an example by Google as the key element in their recent proof of the advantages quantum computing has over traditional computing.
However, dealing with the causes of errors is similar to peeling an onion: Peeling one layer shows the next. In this instance, even with couplers with tunable settings, the two-qubit gates were still susceptible to errors due to unintended interactions among the qubits and between the qubits and coupler. Unwanted interactions like this were typically overlooked before introducing couplers that could be tunable since they were not noticeable until now. Because these residual errors grow as the gates and qubits block the way of developing large-scale quantum processing systems. Physical Review X Physical Review X paper provides an innovative method to minimize these errors.
“We’ve today taken the tunable coupler concept further and demonstrated near 99.9 percent fidelity for both key types of two-qubit gates, called Controlled-Z gates and ISWAP gates,” states William D. Oliver, an associate professor of electrical computer science and engineering, MIT Lincoln Laboratory fellow director of the Center for Quantum Engineering, and director in the Research Laboratory of Electronics, which houses the Executive Quantum Programs group. “higher-fidelity gates raise the number of procedures one may do, and more procedures equals employing more advanced formulas at larger scales.”
To avoid the qubit-qubit interaction, the researchers harnessed the higher energy levels of the coupler to eliminate the undesirable interactions. In the past, these energies of the coupler were not considered even though they caused non-significant qubit-qubit interactions.
“Better controlling and designing the coupler can be an important factor in adjusting the qubit-qubit relationship to the specifications we would like. This is possible by engineering the multilevel dynamic that exists,” Sung says.
The next-generation quantum computers will have error correction, meaning that qubits can be added to enhance the accuracy of quantum computation.
“Qubit errors can be actively addressed by adding redundancy,” Oliver says. Oliver notes that such a procedure is only effective if the gates are adequate. They are at or above a certain level of fidelity dependent on an error-correcting method. “The most flexible thresholds currently are 99 percent or more. In reality, one should seek gate reliability which is more than this to have acceptable levels of hardware redundancy.”
The devices utilized in the study, developed by MIT’s Lincoln Laboratory, were fundamental to the results infidelity and accuracy in the two-qubit operation, Oliver says.
“Fabricating high-coherence devices is step one to implementing high-fidelity control,” the expert says.
Sung Claims, “high rates of error in two-qubit gates considerably restrict the capacity of quantum equipment to perform quantum purposes which are typically difficult to solve with traditional computers, such as, for instance, quantum chemistry simulation and fixing optimization problems.”
At this point, there have been no small molecules that were simulated using quantum computers. Simulations that can be easily executed using conventional computers.
“In this sense, our new approach to reduce the two-qubit gate errors is timely in the field of quantum computation and helps address one of the most critical quantum hardware issues today,” he says.