Abstract | ||
---|---|---|
Most models of decision-making in neuroscience assume an infinite horizon, which yields an optimal solution that integrates evidence up to a fixed decision threshold; however, under most experimental as well as naturalistic behavioral settings, the decision has to be made before some finite deadl ine, which is often experienced as a stochastic quantity, either due to variabl e external constraints or internal timing uncertainty. In this work, we formulate thi s problem as sequential hypothesis testing under a stochastic horizon. We use dynamic programming tools to show that, for a large class of deadline distributions, th e Bayes-optimal solution requires integrating evidence up to a threshold that declin es monotonically over time. We use numerical simulations to illustrate the optimal policy in the special cases of a fixed deadline and one that is drawn from a gamma dist ribution. |
Year | Venue | Keywords |
---|---|---|
2007 | NIPS | numerical simulation,sequential hypothesis testing |
Field | DocType | Citations |
Dynamic programming,Monotonic function,Mathematical optimization,Computer science,Horizon,Infinite horizon,Gamma distribution,Sequential analysis | Conference | 13 |
PageRank | References | Authors |
1.31 | 1 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Peter Frazier | 1 | 61 | 4.99 |
Angela Yu | 2 | 13 | 2.32 |