Merging Quantum Chemistry and Machine Learning for More Accurate Computational Models

Explore a 20-minute Lennard-Jones Centre discussion group seminar by Prof. Geoffrey Hutchison from the University of Pittsburgh that examines how machine learning methods can be combined with quantum chemistry to create more accurate computational models. Learn how traditional computational chemical methods often face a tradeoff between accuracy and speed, and discover how inexpensive approximate quantum chemical methods can be used as components of machine learning models to predict complex properties, including aqueous pKa and organic solar cell performance. Understand how quantum chemical descriptors are calibrated with experimental data through the machine learning model. This Cambridge Materials seminar was held on April 29th, 2024.