Experiments in Value Function Approximation with Sparse Support Vector Regression
Abstract
We present first experiments using Support Vector Regression as function approximator for an on-line, sarsa -like reinforcement learner. To overcome the batch nature of SVR two ideas are employed. The first is sparse greedy approximation: the data is projected onto the subspace spanned by only a small subset of the original data (in feature space). This subset can be built up in an on-line fashion. Second, we use the sparsified data to solve a reduced quadratic problem, where the number of variables is independent of the total number of training samples seen. The feasability of this approach is demonstrated on two common toy-problems.
Cite
Text
Jung and Uthmann. "Experiments in Value Function Approximation with Sparse Support Vector Regression." European Conference on Machine Learning, 2004. doi:10.1007/978-3-540-30115-8_19Markdown
[Jung and Uthmann. "Experiments in Value Function Approximation with Sparse Support Vector Regression." European Conference on Machine Learning, 2004.](https://mlanthology.org/ecmlpkdd/2004/jung2004ecml-experiments/) doi:10.1007/978-3-540-30115-8_19BibTeX
@inproceedings{jung2004ecml-experiments,
title = {{Experiments in Value Function Approximation with Sparse Support Vector Regression}},
author = {Jung, Tobias and Uthmann, Thomas},
booktitle = {European Conference on Machine Learning},
year = {2004},
pages = {180-191},
doi = {10.1007/978-3-540-30115-8_19},
url = {https://mlanthology.org/ecmlpkdd/2004/jung2004ecml-experiments/}
}