| Creators: | Faußer, Stefan A. and Schwenker, Friedhelm |
|---|---|
| Title: | Selective Neural Network Ensembles in Reinforcement Learning |
| Item Type: | Conference or Workshop Item |
| Event Title: | (Proceedings of the) 22nd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN) |
| Event Location: | Bruges, Belgium |
| Event Dates: | 23.-25. April 2014 |
| Page Range: | pp. 105-110 |
| Date: | April 2014 |
| Divisions: | Informationsmanagement |
| Abstract (ENG): | Ensemble models can achieve more accurate predictions than single learners. Selective ensembles further improve the predictions by selecting an informative subset of the full ensemble. We consider rein- forcement learning ensembles, where the members are neural networks. In this context we study a new algorithm for ensemble subset selection in re- inforcement learning scenarios. The aim of the proposed learning strategy is to minimize the Bellman errors of the collected states. In the empirical evaluation, two benchmark applications with large state spaces have been considered, namely SZ-Tetris and generalized maze. Here, our selective ensemble algorithm significantly outperforms other approaches. |
| Forthcoming: | No |
| Language: | English |
| Link eMedia: | Download |
| Citation: | Faußer, Stefan A. and Schwenker, Friedhelm (2014) Selective Neural Network Ensembles in Reinforcement Learning. In: (Proceedings of the) 22nd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), 23.-25. April 2014, Bruges, Belgium, pp. 105-110. ISBN 9782874190957 |
Actions for admins (login required)
 |
View Item in edit mode |