Approximate dynamic programming based on high dimensional model representation
Kybernetika, Tome 49 (2013) no. 5, pp. 720-737
Voir la notice de l'article provenant de la source Czech Digital Mathematics Library
This article introduces an algorithm for implicit High Dimensional Model Representation (HDMR) of the Bellman equation. This approximation technique reduces memory demands of the algorithm considerably. Moreover, we show that HDMR enables fast approximate minimization which is essential for evaluation of the Bellman function. In each time step, the problem of parametrized HDMR minimization is relaxed into trust region problems, all sharing the same matrix. Finding its eigenvalue decomposition, we effectively achieve estimates of all minima. Their full-domain representation is avoided by HDMR and then the same approach is used recursively in the next time step. An illustrative example of N-armed bandit problem is included. We assume that the newly established connection between approximate HDMR minimization and the trust region problem can be beneficial also to many other applications.
Classification :
90C39
Keywords: approximate dynamic programming; Bellman equation; approximate HDMR minimization; trust region problem
Keywords: approximate dynamic programming; Bellman equation; approximate HDMR minimization; trust region problem
@article{KYB_2013__49_5_a3,
author = {Pi\v{s}t\v{e}k, Miroslav},
title = {Approximate dynamic programming based on high dimensional model representation},
journal = {Kybernetika},
pages = {720--737},
publisher = {mathdoc},
volume = {49},
number = {5},
year = {2013},
mrnumber = {3182636},
zbl = {1278.90423},
language = {en},
url = {http://geodesic.mathdoc.fr/item/KYB_2013__49_5_a3/}
}
Pištěk, Miroslav. Approximate dynamic programming based on high dimensional model representation. Kybernetika, Tome 49 (2013) no. 5, pp. 720-737. http://geodesic.mathdoc.fr/item/KYB_2013__49_5_a3/