Q-CTRL, a leader in quantum infrastructure software, has achieved a significant milestone by solving optimization problems over four times larger than previous records using IBM quantum computers. The results show that IBM’s quantum computers can outperform competitive technologies such as quantum annealers and trapped ion devices in solving complex optimization problems. This advancement redefines the expectations for achieving quantum advantage in practical applications.
The new benchmarks demonstrate that Q-CTRL’s performance-management infrastructure, integrated with IBM’s utility-scale quantum computers, effectively mitigates hardware errors, enabling consistent and accurate solutions to large-scale optimization problems. Direct comparisons indicate that using Q-CTRL’s technology, a quantum optimization problem run on a 127-qubit IBM quantum computer was up to 1,500 times more likely to yield the correct result than with an annealer in these specific tests. These tests included two different Max-cut instances and an energy minimization of a 127-node high-order spin-glass model which is an example of a an example of an unconstrained Higher-Order Binary Optimization (HOBO). Q-CTRL achieved these results through the use of a modified QAOA ansatz and update algorithm combined with Q-CTRL’s performance management software for error suppression and mitigation that gets the most out of the available quantum hardware.
There has been skepticism amongst some members of the quantum community that NISQ level gate-based processors without error correction will ever be able to provide a commercial quantum advantage for end users, but this demonstration provides some hope that end users may be able to achieve early quantum advantage without full error correction using these software techniques. Optimization problems may comprise as much as one-third of the overall market for quantum computing applications and these findings suggest significant potential for quantum computing in fields like transport, logistics, machine learning, and financial fraud detection.
For additional information, you can access a press release Q-CTRL has provided that you can see here and a technical paper posted on arXiv here..
June 5, 2024