Convergence of discretization procedure in \(Q\)-learning (Q2725088)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Convergence of discretization procedure in \(Q\)-learning |
scientific article; zbMATH DE number 1618764
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Convergence of discretization procedure in \(Q\)-learning |
scientific article; zbMATH DE number 1618764 |
Statements
21 April 2002
0 references
\(Q\)-learning
0 references
dynamic programming
0 references
discretization
0 references
0 references
0.89569104
0 references
0.8802094
0 references
0.8772295
0 references
0 references
0.8730544
0 references
0.8720821
0 references
Convergence of discretization procedure in \(Q\)-learning (English)
0 references
The authors show that under certain compactness and Lipschitz continuity assumptions, the optimal solution obtained with \(Q\)-learning converges almost surely to the optimal solution obtained with the continuous dynamic programming algorithm as the maximal discretization grids approach zero.
0 references
