Recent advances in accelerated discovery through machine learning and statistical inference

Abstract

Recent applications of machine learning and statistical inference provide case studies demonstrating how such approaches can accelerate the discovery process in physical chemistry and related fields. Examples discussed in this review include the introduction of automated approaches to systematically improve experimental design, increase the efficiency of computationally expensive molecular simulations, facilitate construction of predictive models for complex biological processes, and discover interparticle potentials that lead to materials which meet specified design goals. A common theme is the synergy between experiment and computation enabled by such approaches.

Prepared for Annual Reviews of Physical Chemistry. 24 pages, 7 figures

References in corpus (7)

• Are biological systems poised at criticality?
• Phase diagram of Janus Particles
• A learning scheme to predict atomic forces and accelerate materials simulations
• Phase behavior of a fluid with competing attractive and repulsive interactions
• Turning Statistical Physics Models Into Materials Design Engines
• Genotype to phenotype mapping and the fitness landscape of the E. coli lac promoter
• Inverse spin glass and related maximum entropy problems