Dynamic Planning Networks
| Autor: | Miriam A. M. Capretz, Norman Tasfi |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
reinforcement learning
Traverse Artificial neural network Computer science business.industry Construct (python library) Electrical and Computer Engineering Action (philosophy) deep neural networks Trajectory Reinforcement learning Artificial intelligence State (computer science) Computer Engineering Architecture planning business |
| Zdroj: | Electrical and Computer Engineering Publications IJCNN |
| Popis: | We introduce Dynamic Planning Networks (DPN), a novel architecture for deep reinforcement learning, that combines model-based and model-free aspects for online planning. Our architecture learns to dynamically construct plans using a learned state-transition model by selecting and traversing between simulated states and actions to maximize information before acting. DPN learns to efficiently form plans by expanding a single action-conditional state transition at a time instead of exhaustively evaluating each action, reducing the number of state-transitions used during planning. We observe emergent planning patterns in our agent, including classical search methods such as breadth-first and depth-first search. DPN shows improved performance over existing baselines across multiple axes. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|