Approximate dynamic programming based on high dimensional model representation

Title:

Approximate dynamic programming based on high dimensional model representation

Creator:

Pištěk, Miroslav

Identifier:

https://cdk.lib.cas.cz/client/handle/uuid:fcdd03c6-a825-4e59-b4e1-0b23af5b2374
uuid:fcdd03c6-a825-4e59-b4e1-0b23af5b2374

Subject:

approximate dynamic programming, Bellman equation, approximate HDMR, minimization, and trust region problem

Type:

model:article and TEXT

Format:

bez média and svazek

Description:

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.

Language:

English

Rights:

http://creativecommons.org/publicdomain/mark/1.0/
policy:public

Coverage:

[720]-737

Source:

Kybernetika | 2013 Volume:49 | Number:5

Harvested from:

CDK

Metadata only:

false