Loading...
Please wait, while we are loading the content...
Similar Documents
State Abstraction in MAXQ Hierarchical Reinforcement Learning (2000)
| Content Provider | CiteSeerX |
|---|---|
| Author | Dietterich, Thomas G. |
| Abstract | Many researchers have explored methods for hierarchical reinforcement learning (RL) with temporal abstractions, in which abstract actions are defined that can perform many primitive actions before terminating. However, little is known about learning with state abstractions, in which aspects of the state space are ignored. In previous work, we developed the MAXQ method for hierarchical RL. In this paper, we define five conditions under which state abstraction can be combined with the MAXQ value function decomposition. We prove that the MAXQ-Q learning algorithm converges under these conditions and show experimentally that state abstraction is important for the successful application of MAXQ-Q learning. |
| File Format | |
| Journal | Advances in Neural Information Processing Systems 12 |
| Publisher Date | 2000-01-01 |
| Access Restriction | Open |
| Subject Keyword | State Abstraction Many Researcher Abstract Action State Space Temporal Abstraction Many Primitive Action Maxq Value Function Decomposition Maxq-q Learning Successful Application Hierarchical Reinforcement Learning Maxq Method Maxq Hierarchical Reinforcement Learning Hierarchical Rl Maxq-q Learning Algorithm Converges |
| Content Type | Text |
