CaltechAUTHORS
  A Caltech Library Service

Evolutionary reinforcement learning of dynamical large deviations

Whitelam, Stephen and Jacobson, Daniel and Tamblyn, Isaac (2020) Evolutionary reinforcement learning of dynamical large deviations. Journal of Chemical Physics, 153 (4). Art. No. 044113. ISSN 0021-9606. https://resolver.caltech.edu/CaltechAUTHORS:20200728-141349560

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200728-141349560

Abstract

We show how to bound and calculate the likelihood of dynamical large deviations using evolutionary reinforcement learning. An agent, a stochastic model, propagates a continuous-time Monte Carlo trajectory and receives a reward conditioned upon the values of certain path-extensive quantities. Evolution produces progressively fitter agents, potentially allowing the calculation of a piece of a large-deviation rate function for a particular model and path-extensive quantity. For models with small state spaces, the evolutionary process acts directly on rates, and for models with large state spaces, the process acts on the weights of a neural network that parameterizes the model’s rates. This approach shows how path-extensive physics problems can be considered within a framework widely used in machine learning.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1063/5.0015301DOIArticle
ORCID:
AuthorORCID
Whitelam, Stephen0000-0002-0086-6803
Tamblyn, Isaac0000-0002-8146-6667
Additional Information:© 2020 Published under license by AIP Publishing. Submitted: 27 May 2020; Accepted: 31 May 2020; Published Online: 27 July 2020. We thank Hugo Touchette for the comments. This work was performed as part of a user project at the Molecular Foundry, Lawrence Berkeley National Laboratory, supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02–05CH11231. D.J. acknowledges support from the Department of Energy Computational Science Graduate Fellowship. I.T. performed work at the National Research Council of Canada under the auspices of the AI4D Program. D.J. acknowledges support from the Department of Energy Computational Science Graduate Fellowship, under Contract No. DE-FG02-97ER25308.
Funders:
Funding AgencyGrant Number
Department of Energy (DOE)DE-AC02-05CH11231
Department of Energy (DOE)DE-FG02-97ER25308
Issue or Number:4
Record Number:CaltechAUTHORS:20200728-141349560
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200728-141349560
Official Citation:Evolutionary reinforcement learning of dynamical large deviations. Stephen Whitelam, Daniel Jacobson, Isaac Tamblyn. The Journal of Chemical Physics 153:4; doi: 10.1063/5.0015301
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:104610
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:28 Jul 2020 22:44
Last Modified:28 Jul 2020 22:44

Repository Staff Only: item control page