Spin glasses are disordered magnets with random interactions that are, generally, in conflict with each other. Finding the ground states of spin glasses is not only essential for understanding the nature of disordered magnets and many other physical systems, but also useful to solve a broad array of hard combinatorial optimization problems across multiple disciplines. Despite decades-long efforts, an algorithm with both high accuracy and high efficiency is still lacking. Here we introduce DIRAC – a deep reinforcement learning framework, which can be trained purely on small-scale spin glass instances and then applied to arbitrarily large ones. DIRAC displays better scalability than other methods and can be leveraged to enhance any thermal annealing method. Extensive calculations on 2D, 3D and 4D Edwards-Anderson spin glass instances demonstrate the superior performance of DIRAC over existing methods. The presented framework will help us better understand the nature of the low-temperature spin-glass phase, which is a fundamental challenge in statistical physics. Moreover, the gauge transformation technique adopted in DIRAC builds a deep connection between physics and artificial intelligence. In particular, this opens up a promising avenue for reinforcement learning models to explore in the enormous configuration space, which would be extremely helpful to solve many other hard combinatorial optimization problems.

Ref: https://www.nature.com/articles/s41467-023-36363-w

Biography:

Dr. Yang-Yu Liu is currently an Associate Professor of Medicine at Harvard Medical School (HMS) and an Associate Scientist at Brigham and Women’s Hospital (BWH). He received his PhD in Physics from University of Illinois at Urbana-Champaignin 2009. After that, he held positions as Postdoctoral Research Associate and then Research Assistant Professor in the Center for Complex Network Research at Northeastern University, before he joined HMS and BWH in 2013. His current research efforts focus on the study of human microbiome from the community ecology, dynamical systems, control theory, and machine learning perspectives. For more information, please visit http://scholar.harvard.edu/yyl/

所内联系人： 张潘