Scalable transfer learning in heterogeneous, dynamic environments

Reinforcement learning is a plausible theoretical basis for developing self-learning, autonomous agents or robots that can effectively represent the world dynamics and efficiently learn the problem features to perform different tasks in different environments. The computational costs and complexitie...

Full description

Saved in:
Bibliographic Details
Main Authors: Nguyen, Trung Thanh, Silander, Tomi, LI, Zhuoru, Tze-Yun LEONG
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2017
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/3039
https://ink.library.smu.edu.sg/context/sis_research/article/4039/viewcontent/ScalableTransferLearningHeterogeneousDynamicEnvironments_2015.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-4039
record_format dspace
spelling sg-smu-ink.sis_research-40392020-04-02T06:23:50Z Scalable transfer learning in heterogeneous, dynamic environments Nguyen, Trung Thanh Silander, Tomi LI, Zhuoru Tze-Yun LEONG, Reinforcement learning is a plausible theoretical basis for developing self-learning, autonomous agents or robots that can effectively represent the world dynamics and efficiently learn the problem features to perform different tasks in different environments. The computational costs and complexities involved, however, are often prohibitive for real-world applications. This study introduces a scalable methodology to learn and transfer knowledge of the transition (and reward) models for model-based reinforcement learning in a complex world. We propose a variant formulation of Markov decision processes that supports efficient online-learning of the relevant problem features to approximate the world dynamics. We apply the new feature selection and dynamics approximation techniques in heterogeneous transfer learning, where the agent automatically maintains and adapts multiple representations of the world to cope with the different environments it encounters during its lifetime. We prove regret bounds for our approach, and empirically demonstrate its capability to quickly converge to a near optimal policy in both real and simulated environments. 2017-06-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/3039 info:doi/10.1016/j.artint.2015.09.013 https://ink.library.smu.edu.sg/context/sis_research/article/4039/viewcontent/ScalableTransferLearningHeterogeneousDynamicEnvironments_2015.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Model-based reinforcement learning Online feature selection Transfer learning Artificial Intelligence and Robotics
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Model-based reinforcement learning
Online feature selection
Transfer learning
Artificial Intelligence and Robotics
spellingShingle Model-based reinforcement learning
Online feature selection
Transfer learning
Artificial Intelligence and Robotics
Nguyen, Trung Thanh
Silander, Tomi
LI, Zhuoru
Tze-Yun LEONG,
Scalable transfer learning in heterogeneous, dynamic environments
description Reinforcement learning is a plausible theoretical basis for developing self-learning, autonomous agents or robots that can effectively represent the world dynamics and efficiently learn the problem features to perform different tasks in different environments. The computational costs and complexities involved, however, are often prohibitive for real-world applications. This study introduces a scalable methodology to learn and transfer knowledge of the transition (and reward) models for model-based reinforcement learning in a complex world. We propose a variant formulation of Markov decision processes that supports efficient online-learning of the relevant problem features to approximate the world dynamics. We apply the new feature selection and dynamics approximation techniques in heterogeneous transfer learning, where the agent automatically maintains and adapts multiple representations of the world to cope with the different environments it encounters during its lifetime. We prove regret bounds for our approach, and empirically demonstrate its capability to quickly converge to a near optimal policy in both real and simulated environments.
format text
author Nguyen, Trung Thanh
Silander, Tomi
LI, Zhuoru
Tze-Yun LEONG,
author_facet Nguyen, Trung Thanh
Silander, Tomi
LI, Zhuoru
Tze-Yun LEONG,
author_sort Nguyen, Trung Thanh
title Scalable transfer learning in heterogeneous, dynamic environments
title_short Scalable transfer learning in heterogeneous, dynamic environments
title_full Scalable transfer learning in heterogeneous, dynamic environments
title_fullStr Scalable transfer learning in heterogeneous, dynamic environments
title_full_unstemmed Scalable transfer learning in heterogeneous, dynamic environments
title_sort scalable transfer learning in heterogeneous, dynamic environments
publisher Institutional Knowledge at Singapore Management University
publishDate 2017
url https://ink.library.smu.edu.sg/sis_research/3039
https://ink.library.smu.edu.sg/context/sis_research/article/4039/viewcontent/ScalableTransferLearningHeterogeneousDynamicEnvironments_2015.pdf
_version_ 1770572787834945536