Machine learning in condensed matter physics: recent advances and opportunities*

With the rapid advances in the machine learning technology in recent years, we are witnessing exploding interests in its applications to solve physics problems. Pioneering work has proven machine learning a promising tool to tackle many frontier research problems in condensed matter physics, such as the identification of phases of matter, solving quantum many-body problems, accelerating materials-by-design, and decreasing computational complexity for simulations, to name a few. Machine learning opens up new research avenues that bridge across theory, experiments and computer simulations, and it enables scientific discoveries that were impossible before. In this talk, I will give a brief overview of the current state-of-the-art, as well as opportunities, in this exciting, emerging area.