Roadmap on Machine learning in electronic structure

Abstract In recent years, we have been witnessing a paradigm shift in computational materials science. fact, traditional methods, mostly developed the second half of XXth century, are being complemented, extended, and sometimes even completely replaced by faster, simpler, often more accurate approaches. The new approaches, that collectively label machine learning, their origins fields informatics artificial intelligence, but making rapid inroads all other branches With this mind, Roadmap article, consisting multiple contributions from experts across field, discusses use learning science, share perspectives on current future challenges problems as diverse prediction properties, construction force-fields, development exchange correlation functionals for density-functional theory, solution many-body problem, more. spite already numerous exciting success stories, just at beginning long path will reshape science many XXIth century.