literature
Literature
Authors
Jobs
Seminars
Conferences
DataBETA

Reinforcement learning approach to shortcuts between thermodynamic states with minimum entropy production

Nov 5, 2021
13 pages
Published in:
  • Phys.Rev.E 105 (2022) 5, 054123
  • Published: May 13, 2022
e-Print:
DOI:
View in:
pdf
reference search2 citations

Citations per year

Loading ...
Abstract: (APS)
We propose a systematic method based on reinforcement learning (RL) techniques to find the optimal path to minimize the total entropy production between two equilibrium states of open systems at the same temperature in a given fixed period. Benefiting from the generalization of the deep RL techniques, we provide a powerful tool to address this problem in quantum systems even with two-dimensional continuous controllable parameters. We successfully apply our method to the classical and quantum two-level systems.
Note:
  • 13 pages, 4 figures
Loading ...