The Governors of the University of Calgary

Published purpose

Current efforts to design materials computationally face two bottlenecks: accuracy and speed. The workhorse density-functional theory (DFT) computational methods are time-consuming and of limited accuracy for strongly correlated systems like the metal-oxides commonly used for catalysis. (Classical) Machine learning (ML) is showing great promise in reducing the number of expensive DFT calculations in the design process. Quantum computing (focusing on hybrid approaches using Noisy Intermediate Scale Quantum (NISQ) computers for parts of the problem) is showing great promise to contribute to opening both bottlenecks by facilitating the use of more accurate quantum chemical methods and accelerating machine-learning approaches beyond the possibilities of classical ML. The proposed research will be directed along two complementary streams: i) quantum computing for quantum chemistry and ii) quantum machine learning.