The learning objectives will be to provide an introduction into state of the art for machine learning in materials science. The course will be a whirlwind of current applications of machine learning and deep learning to areas in theory, where ML can be highly beneficial for learning force fields, as well as property prediction to accelerate computational searches for new materials, as well as use of ML/DL in imaging and spectral processing settings for understanding and extracting physics from multidimensional datasets, including methods to incorporate prior knowledge to accelerate learning. In addition, there will be an opportunity to learn how Bayesian optimization and reinforcement learning can be applied within an experimental regime to facilitate smart experiments, guided by optimal parameter selections.

The morning session will cover:

• ML to theory (force fields, generic property predictions, visualization, accessing the different databases, etc.)
• Deep learning introduction and use to segment images from electron or scanning probe microscopy

The afternoon session will cover:

• Spectral processing with matrix/tensor factorization and deep learning methods (including autoencoders) for extracting data from large multidimensional datasets
• Reinforcement learning and Bayesian optimization for efficient and smart experiments