Networks with input gates for situation-dependent input selection in reinforcement learning
説明
A method is proposed for situation-dependent input selection and learning acceleration in Q-learning. Q-values are expressed by an RBF network which has an input gate attached to each of its input channels in order to capture, by learning, situation-dependent relevance or usefulness of the input.
収録刊行物
-
- Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)
-
Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290) 5-10, 2003-06-25
IEEE