Transferring instances for model-based reinforcement learning

Transferring instances for model-based reinforcement learning (2008)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Matthew E. Taylor , Nicholas K. Jong , Peter Stone

Venue:	AAMAS 2008 Workshop on Adaptive Learning Agents and Multi-Agent Systems
Citations:	17 - 3 self

BibTeX

@INPROCEEDINGS{Taylor08transferringinstances,
    author = {Matthew E. Taylor and Nicholas K. Jong and Peter Stone},
    title = {Transferring instances for model-based reinforcement learning},
    booktitle = {AAMAS 2008 Workshop on Adaptive Learning Agents and Multi-Agent Systems},
    year = {2008}
}

Bookmark

OpenURL

Abstract

Reinforcement learning agents typically require a significant amount of data before performing well on complex tasks. Transfer learning methods have made progress reducing sample complexity, but they have only been applied to model-free learning methods, not more data-efficient model-based learning methods. This paper introduces TIMBREL, a novel method capable of transferring information effectively into a model-based reinforcement learning algorithm. We demonstrate that TIMBREL can significantly improve the sample complexity and asymptotic performance of a model-based algorithm when learning in a continuous state space.

Citations

1965	Dynamic Programming - Bellman - 1957
1135	Learning from delayed rewards - Watkins - 1989
878	Markov Decision Processes: Discrete Stochastic Dynamic Programming - Puterman - 1994
649	Reinforcement learning - Sutton, Barto - 1998
329	Multitask learning - Caruana - 1997
275	Prioritized sweeping: Reinforcement learning with less data and less time - Moore, Atkeson - 1993
250	Improving elevator performance using reinforcement learning - Crites, Barto - 1996
224	On-line Q-learning using connectionist systems - Rummery, Niranjan - 1994
190	Td-gammon, a self-teaching backgammon program, achieves master-level play - Tesauro - 1994
176	R-max: A general polynomial time algorithm for near-optimal reinforcement learning - Brafman, Tennenholtz - 2001
174	Near-optimal reinforcement learning in polynomial time - Kearns, Singh - 1998
168	PEGASUS: A policy search method for large MDPs and POMDPs - Ng, Jordan - 2000
166	Reinforcement learning with replacing eligibility traces - Singh, Sutton - 1996
135	M.: Learning in the presence of concept drift and hidden contexts - Widmer, Kubat - 1996
106	Autonomous helicopter flight via reinforcement learning - Ng, Kim, et al.
102	A reinforcement learning approach to job-shop scheduling - Zhang, Dietterich - 1995
85	Reinforcement Learning for RoboCupsoccer Keepaway - Stone, Sutton, et al. - 2005
69	Machine learning for fast quadrupedal locomotion - Kohl, Stone - 2004
67	Model based Bayesian exploration - Dearden, Friedman, et al. - 1999
34	Using advice to transfer knowledge acquired in one Reinforcement Learning task to another - Torrey, Walker, et al. - 2005
33	Using homomorphisms to transfer options across continuous reinforcement learning domains - Soni, Singh - 2006
30	A comparison of direct and model-based reinforcement learning - Atkeson, Santamaria - 1997
23	Multi-task reinforcement learning: a hierarchical Bayesian approach - Wilson, Fern, et al.
22	Transfer learning via inter-task mappings for temporal difference learning - Taylor, Stone, et al.
22	Transfer via inter-task mappings in policy search reinforcement learning - Taylor, Whiteson, et al. - 2007
21	Value-function-based transfer for reinforcement learning using structure mapping - Liu, Stone - 2006
13	Model-based exploration in continuous state spaces - Jong, Stone - 2007
13	Autonomous Transfer for Reinforcement Learning - Taylor, Kuhlmann, et al. - 2008
12	Transfer of samples in batch reinforcement learning - Lazarich, Restelli, et al.
9	Autonomous learning of stable quadruped locomotion - Saggar, D’Silva, et al.
8	Transfer Learning in Reinforcement Learning Problems through Partial Policy Recycling - Ramon, Driessens, et al. - 2007
7	Multitask reinforcement learning on the distribution of MDPs - Tanaka, Yamamura
5	Model transfer for Markov decision tasks via parameter matching - Sunmola, Wyatt