Prefer a chat interface with context about you and your work?
Reinforcement Learning Approach to Nonequilibrium Quantum Thermodynamics
We use a reinforcement learning approach to reduce entropy production in a closed quantum system brought out of equilibrium. Our strategy makes use of an external control Hamiltonian and a policy gradient technique. Our approach bears no dependence on the quantitative tool chosen to characterize the degree of thermodynamic irreversibility …