Abstract
Abstract:
We consider conditions that allow us to find an optimal strategy for sequential decisions from a given data situation. For the case where all interventions are unconditional (atomic), identifiability has been discussed by Pearl & Robins (1995). We argue here that an optimal strategy must be conditional, i.e. take the information available at each decision point into account. We show that the identification of an optimal sequential decision strategy is more restrictive, in the sense that conditional interventions might not always be identified when atomic interventions are. We further demonstrate that a simple graphical criterion for the identifiability of an optimal strategy can be given.
Translated title of the contribution | Identifying optimal sequential decisions |
---|---|
Original language | English |
Pages | 113 - 120 |
Number of pages | 8 |
Publication status | Published - 2008 |
Event | 24th Annual Conference on Uncertainty in Artifical Intelligence - Helsinki, Finland Duration: 9 Jul 2008 → … |
Conference
Conference | 24th Annual Conference on Uncertainty in Artifical Intelligence |
---|---|
Country/Territory | Finland |
City | Helsinki |
Period | 9/07/08 → … |
Bibliographical note
Name and Venue of Event: 24th UAI Conference, Helsinki, FinlandConference Proceedings/Title of Journal: Proceedings of the 24th UAI Conference
Conference Organiser: Ronald Parr