Daniel Lidar is a professor of electrical engineering, chemistry, and physics at the Viterbi School of Engineering. His main research interest is quantum information processing, specifically quantum control, quantum error correction, the theory of open quantum systems, quantum algorithms, and theoretical as well as experimental adiabatic quantum computation.
Lidar and his team work on illuminating the computational power of quantum computers. Quantum computers, using the laws of quantum mechanics, are able to process vast numbers of calculations at the same time, making them significantly more powerful than traditional computers. Quantum computers can efficiently simulate other quantum systems and predict their properties. Since large-scale quantum computers do not currently exist, Lidar and his team must simulate them on classical computers, a task that involves performing thousands of simulations on the HPC clusters.
Performing these simulations lets researchers test how quantum computers will act under real-life, noisy conditions, helping them develop error-correction methods for overcoming the adverse effects of noise caused by heat. Lidar and his team have recently developed such an error-correction method to protect a quantum optimizer, a special type of quantum computer built by D-Wave Systems and housed at the USC-Lockheed Martin Quantum Computing Center.
Lidar’s team discovered that coupling several quantum bits (qubits) together on a D-Wave Two quantum optimizer, without changing the hardware of the device, will cause the qubits to act effectively as one single qubit that experiences a lower temperature. The more qubits are coupled, the more the temperature decreases , allowing researchers to minimize the effects of noise resulting from heat. This error correction scheme is implementable not only on platforms such as the D-Wave processor on which it was tested, but also on future quantum optimization devices with different hardware architectures. To ensure a complete understanding of the mechanisms underlying the error correction scheme, the team performed thousands of simulations on the HPC cluster. The work was published in the journal Nature Quantum Information, vol.2, p.16017 (2016).
ABOVE: An illustration of the error correction scheme. Each circle represents a physical qubit, and groups of the same color represent an encoded qubit that is later decoded as part of the error correction. Thick red lines represent “energy penalties” that keep the qubits aligned and thus more resistant to errors. Thin lines represent logical couplings of varying strengths that express an optimization problem the quantum optimizer has to solve.