Finding the ground state of spin Hamiltonians with reinforcement learning

Reinforcement learning has become a popular method in various domains, for problems where an agent must learn what actions must be taken to reach a particular goal. An interesting example where the technique can be applied is simulated annealing in condensed matter physics, where a procedure is determined for slowly cooling a complex system to its ground state. A reinforcement learning approach has been developed that can learn a temperature scheduling protocol to find the ground state of spin glasses, magnetic systems with strong spin-spin interactions between neighbouring atoms. Reinforcement learning (RL) has become a proven method for optimizing a procedure for which success has been defined, but the specific actions needed to achieve it have not. Using a method we call 'controlled online optimization learning' (COOL), we apply the so-called 'black box' method of RL to simulated annealing (SA), demonstrating that an RL agent based on proximal policy optimization can, through experience alone, arrive at a temperature schedule that surpasses the performance of standard heuristic temperature schedules for two classes of Hamiltonians. When the system is initialized at a cool temperature, the RL agent learns to heat the system to 'melt' it and then slowly cool it in an effort to anneal to the ground state; if the system is initialized at a high temperature, the algorithm immediately cools the system. We investigate the performance of our RL-driven SA agent in generalizing to all Hamiltonians of a specific class. When trained on random Hamiltonians of nearest-neighbour spin glasses, the RL agent is able to control the SA process for other Hamiltonians, reaching the ground state with a higher probability than a simple linear annealing schedule. Furthermore, the scaling performance (with respect to system size) of the RL approach is far more favourable, achieving a performance improvement of almost two orders of magnitude onL= 14(2)systems. We demonstrate the robustness of the RL approach when the system operates in a 'destructive observation' mode, an allusion to a quantum system where measurements destroy the state of the system. The success of the RL agent could have far-reaching impacts, from classical optimization, to quantum annealing and to the simulation of physical systems.