skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning

Abstract

We assume that we are given a time series of data from a dynamical system and our task is to learn the flow map of the dynamical system. We present a collection of results on how to enforce constraints coming from the dynamical system in order to accelerate the training of deep neural networks to represent the flow map of the system as well as increase their predictive ability. In particular, we provide ways to enforce constraints during training for all three major modes of learning, namely supervised, unsupervised and reinforcement learning. In general, the dynamic constraints need to include terms which are analogous to memory terms in model reduction formalisms. Such memory terms act as a restoring force which corrects the errors committed by the learned flow map during prediction. For supervised learning, the constraints are added to the objective function. For the case of unsupervised learning, in particular generative adversarial networks, the constraints are introduced by augmenting the input of the discriminator. Finally, for the case of reinforcement learning and in particular actor-critic methods, the constraints are added to the reward function. In addition, for the reinforcement learning case, we present a novel approach based on homotopy ofmore » the action-value function in order to stabilize and accelerate training. We use numerical results for the Lorenz system to illustrate the various constructions.« less

Authors:
 [1]
  1. BATTELLE (PACIFIC NW LAB)
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1631321
Report Number(s):
PNNL-SA-143654
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: Proceedings of the AAAI 2020 Spring Symposium on Combining Artificial Intelligence and Machine Learning with Physical Sciences (AAAI-MLPS 2020), March 23-25, 2020, Stanford, CA
Country of Publication:
Germany
Language:
English

Citation Formats

Stinis, Panagiotis. Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning. Germany: N. p., 2020. Web.
Stinis, Panagiotis. Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning. Germany.
Stinis, Panagiotis. 2020. "Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning". Germany.
@article{osti_1631321,
title = {Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning},
author = {Stinis, Panagiotis},
abstractNote = {We assume that we are given a time series of data from a dynamical system and our task is to learn the flow map of the dynamical system. We present a collection of results on how to enforce constraints coming from the dynamical system in order to accelerate the training of deep neural networks to represent the flow map of the system as well as increase their predictive ability. In particular, we provide ways to enforce constraints during training for all three major modes of learning, namely supervised, unsupervised and reinforcement learning. In general, the dynamic constraints need to include terms which are analogous to memory terms in model reduction formalisms. Such memory terms act as a restoring force which corrects the errors committed by the learned flow map during prediction. For supervised learning, the constraints are added to the objective function. For the case of unsupervised learning, in particular generative adversarial networks, the constraints are introduced by augmenting the input of the discriminator. Finally, for the case of reinforcement learning and in particular actor-critic methods, the constraints are added to the reward function. In addition, for the reinforcement learning case, we present a novel approach based on homotopy of the action-value function in order to stabilize and accelerate training. We use numerical results for the Lorenz system to illustrate the various constructions.},
doi = {},
url = {https://www.osti.gov/biblio/1631321}, journal = {},
number = ,
volume = ,
place = {Germany},
year = {Fri Apr 03 00:00:00 EDT 2020},
month = {Fri Apr 03 00:00:00 EDT 2020}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: