Temporal Difference Networks for Dynamical Systems with Continuous Observations and Actions

Christopher M. Vigorito
Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, PMLR R7:83-90, 2009.

Abstract

Temporal-difference (TD) networks are a class of predictive state representations that use well-established TD methods to learn models of partially observable dynamical systems. Previous research with TD networks has dealt only with dynamical systems with finite sets of observations and actions. We present an algorithm for learning TD network representations of dynamical systems with continuous observations and actions. Our results show that the algorithm is capable of learning accurate and robust models of several noisy continuous dynamical systems. The algorithm presented here is the first fully incremental method for learning a predictive representation of a continuous dynamical system.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR7-vigorito09a, title = {Temporal Difference Networks for Dynamical Systems with Continuous Observations and Actions}, author = {Vigorito, Christopher M.}, booktitle = {Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence}, pages = {83--90}, year = {2009}, editor = {Bilmes, Jeff and Ng, Andrew Y.}, volume = {R7}, series = {Proceedings of Machine Learning Research}, month = {18--21 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r7/main/assets/vigorito09a/vigorito09a.pdf}, url = {https://proceedings.mlr.press/r7/vigorito09a.html}, abstract = {Temporal-difference (TD) networks are a class of predictive state representations that use well-established TD methods to learn models of partially observable dynamical systems. Previous research with TD networks has dealt only with dynamical systems with finite sets of observations and actions. We present an algorithm for learning TD network representations of dynamical systems with continuous observations and actions. Our results show that the algorithm is capable of learning accurate and robust models of several noisy continuous dynamical systems. The algorithm presented here is the first fully incremental method for learning a predictive representation of a continuous dynamical system.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Temporal Difference Networks for Dynamical Systems with Continuous Observations and Actions %A Christopher M. Vigorito %B Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2009 %E Jeff Bilmes %E Andrew Y. Ng %F pmlr-vR7-vigorito09a %I PMLR %P 83--90 %U https://proceedings.mlr.press/r7/vigorito09a.html %V R7 %X Temporal-difference (TD) networks are a class of predictive state representations that use well-established TD methods to learn models of partially observable dynamical systems. Previous research with TD networks has dealt only with dynamical systems with finite sets of observations and actions. We present an algorithm for learning TD network representations of dynamical systems with continuous observations and actions. Our results show that the algorithm is capable of learning accurate and robust models of several noisy continuous dynamical systems. The algorithm presented here is the first fully incremental method for learning a predictive representation of a continuous dynamical system. %Z Reissued by PMLR on 04 October 2026.
APA
Vigorito, C.M.. (2009). Temporal Difference Networks for Dynamical Systems with Continuous Observations and Actions. Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R7:83-90 Available from https://proceedings.mlr.press/r7/vigorito09a.html. Reissued by PMLR on 04 October 2026.

Related Material