Multi-View Decision Processes: The Helper-AI Problem
MetadataShow full item record
CitationDimitrakakis, Christos, David C. Parkes, Goran Radanovic, and Paul Tylkin. 2017. Multi-View Decision Processes: The Helper-AI Problem. 31st Conference on Neural Information Processing Systems (NIPS).
AbstractWe consider a two-player sequential game in which agents have the same reward function but may disagree on the transition probabilities of an underlying Markovian model of the world. By committing to play a specific policy, the agent with the correct model can steer the behavior of the other agent, and seek to improve utility. We model this setting as a multi-view decision process, which we use to formally analyze the positive effect of steering policies. Furthermore, we develop an algorithm for computing the agents' achievable joint policy, and we experimentally show that it can lead to a large utility increase when the agents' models diverge.
Citable link to this pagehttp://nrs.harvard.edu/urn-3:HUL.InstRepos:41302220
- FAS Scholarly Articles 
Contact administrator regarding this item (to report mistakes or request changes)