Machine Learning in Condensed Matter Physics

Machine learning is becoming an important tool in condensed matter physics, opening new avenues for the study of complex quantum and/or many-body systems. As data-driven science emerges as a new paradigm of scientific research, this volume brings together leading experts to examine the role of machine learning across a broad range of problems in the field. Beyond presenting recent advances, the book discusses common opportunities and challenges that cut across diverse applications, including trainability, interpretability, inverse problems, and the search for new physical insights. The chapters cover a broad spectrum of topics, including quantum phase recognition and phase transitions, quantum many-body problems, quantum chemistry, generative models, neural quantum states, self-energy spectroscopy, materials informatics, density functional theory, self-learning Monte Carlo methods, and quantum machine learning. Each contribution reviews the current state of the art while discussing the significance, challenges, and future directions of machine learning in its respective area. Serving as both an introduction and a reference, this book is of interest to graduate students and researchers in condensed matter physics, materials science, computational physics, and quantum information science. By connecting research directions that have often evolved independently, it offers a coherent perspective on the relationship between machine learning and the physics of quantum matter.