Loading...
Please wait, while we are loading the content...
Similar Documents
Approximate solutions of interactive pomdps using point based value iteration.
| Content Provider | CiteSeerX |
|---|---|
| Author | Perez, Dennis D. |
| Abstract | We develop a point based method for solving finitely nested interactive POMDPs approximately. Analogously to point based value iteration (PBVI) in POMDPs, we maintain a set of belief points and form value functions composed of only those value vectors that are optimal at these points. However, as we focus on multiagent settings, the beliefs are nested and the computation of the value vectors relies on predicted actions of others. Consequently, we develop an interactive generalization of PBVI applicable to multiagent settings. We bound the error theoretically and provide empirical results using multiple domains. 1 |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Interactive Generalization Multiagent Setting Value Vector Interactive Pomdps Using Point Belief Point Interactive Pomdps Approximate Solution Value Vector Relies Multiple Domain Form Value Function Value Iteration Predicted Action Empirical Result |
| Content Type | Text |
