Abstract
Mechanisms are proposed for updating beliefs at nodes in probabilistic inference networks, taking into account the time points associated with observations and decisions. Simple models with few parameters are proposed, addressing dependence on exact time points as well as time delays before final belief computation. These models are general enough to accommodate applications whose characteristics differ with respect to how the reliability of information ages with time.
Original language | English (US) |
---|---|
Pages | 93-100 |
Number of pages | 8 |
State | Published - 2001 |
Event | 2001 AAAI Fall Symposium - North Falmouth, United States Duration: Nov 2 2001 → Nov 4 2001 |
Conference
Conference | 2001 AAAI Fall Symposium |
---|---|
Country/Territory | United States |
City | North Falmouth |
Period | 11/2/01 → 11/4/01 |
ASJC Scopus subject areas
- General Engineering