On-Line Discovery of Temporal-Difference Networks
Abstract
We present an algorithm for on-line, incremental discovery of temporal-difference (TD) networks. The key contribution is the establishment of three criteria to expand a node in TD network: a node is expanded when the node is well-known, independent, and has a prediction error that requires further explanation. Since none of these criteria requires centralized calculation operations, they are easily computed in a parallel and distributed manner, and scalable for bigger problems compared to other discovery methods of predictive state representations. Through computer experiments, we demonstrate the empirical effectiveness of our algorithm.
Cite
Text
Makino and Takagi. "On-Line Discovery of Temporal-Difference Networks." International Conference on Machine Learning, 2008. doi:10.1145/1390156.1390236Markdown
[Makino and Takagi. "On-Line Discovery of Temporal-Difference Networks." International Conference on Machine Learning, 2008.](https://mlanthology.org/icml/2008/makino2008icml-line/) doi:10.1145/1390156.1390236BibTeX
@inproceedings{makino2008icml-line,
title = {{On-Line Discovery of Temporal-Difference Networks}},
author = {Makino, Takaki and Takagi, Toshihisa},
booktitle = {International Conference on Machine Learning},
year = {2008},
pages = {632-639},
doi = {10.1145/1390156.1390236},
url = {https://mlanthology.org/icml/2008/makino2008icml-line/}
}