The scientific process of discovering new knowledge is often characterized as search from a space of candidates, and machine learning can accelerate the search by properly modeling the data and suggesting which candidates to apply experiments on. In many cases, experiments can be substituted by first principles calculation. I review two basic machine learning techniques called Bayesian optimization and Monte Carlo tree search. I also show successful case studies including Si-Ge nanostructure design, optimization of grain boundary structures and discovery of low-thermal-conductivity compounds from a database.

Reference:
S. Ju, T. Shiga, L. Feng, Z. Hou, K. Tsuda, and J. Shiomi, Designing Nanostructures for Phonon Transport via Bayesian Optimization, Physical Review X, 7, 021024, 2017.

S. Kiyohara, H. Oda, K. Tsuda and T. Mizoguchi, Acceleration of stable interface structure searching using a kriging approach, Japanese Journal of Applied Physics, 55, 045502, 2016.

A. Seko, A. Togo, H. Hayashi, K. Tsuda, L. Chaput, and I. Tanaka, Prediction of low-thermal-conductivity compounds with first-principles anharmonic lattice-dynamics calculations and Bayesian optimization, Physical Review Letters, 115, 205901, 2015.