Number of the records: 1

Approximate Dynamic Programming Based on High Dimensional Model Representation

SYSNO ASEP	0399560
Document Type	J - Journal Article
R&D Document Type	Journal Article
Subsidiary J	Článek ve WOS
Title	Approximate Dynamic Programming Based on High Dimensional Model Representation
Author(s)	Pištěk, Miroslav (UTIA-B)_{RID, ORCID}
Source Title	Kybernetika. - : Ústav teorie informace a automatizace AV ČR, v. v. i. - ISSN 0023-5954 Roč. 49, č. 5 (2013), s. 720-737
Number of pages	18 s.
Publication form	Print - P
Language	eng - English
Country	CZ - Czech Republic
Keywords	approximate dynamic programming ; Bellman equation ; approximate HDMR minimization ; trust region problem
Subject RIV	BC - Control Systems Theory
R&D Projects	GAP102/11/0437 GA ČR - Czech Science Foundation (CSF)
Institutional support	UTIA-B - RVO:67985556
UT WOS	000328665200004
EID SCOPUS	84889006605
Annotation	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 min- imization 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.
Workplace	Institute of Information Theory and Automation
Contact	Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201.
Year of Publishing	2014

Number of the records: 1