Transferring Expectations in Model-Based Reinforcement Learning
Abstract
We study how to automatically select and adapt multiple abstractions or representations of the world to support model-based reinforcement learning. We address the challenges of transfer learning in heterogeneous environments with varying tasks. We present an efficient, online framework that, through a sequence of tasks, learns a set of relevant representations to be used in future tasks. Without pre-defined mapping strategies, we introduce a general approach to support transfer learning across different state spaces. We demonstrate the potential impact of our system through improved jumpstart and faster convergence to near optimum policy in two benchmark domains.
Cite
Text
Nguyen et al. "Transferring Expectations in Model-Based Reinforcement Learning." Neural Information Processing Systems, 2012.Markdown
[Nguyen et al. "Transferring Expectations in Model-Based Reinforcement Learning." Neural Information Processing Systems, 2012.](https://mlanthology.org/neurips/2012/nguyen2012neurips-transferring/)BibTeX
@inproceedings{nguyen2012neurips-transferring,
title = {{Transferring Expectations in Model-Based Reinforcement Learning}},
author = {Nguyen, Trung and Silander, Tomi and Leong, Tze Y.},
booktitle = {Neural Information Processing Systems},
year = {2012},
pages = {2555-2563},
url = {https://mlanthology.org/neurips/2012/nguyen2012neurips-transferring/}
}