2013/01/10 by Paat Rusmevichientong, Rusmevichientong, Paat, Benjamin Van Roy +2
Computer Science · Decision Sciences · #Artificial Intelligence (cs.AI) #Auction Theory and Applications #Bayesian Modeling and Causal Inference #FOS: Computer and information sciences #Reinforcement Learning in Robotics #cs.AI
paper · pdf · doi:10.48550/arxiv.1301.2308
Appears in Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (UAI2001)
arxiv created 2013/01/10 · openalex publication_date 2013/01/10 · arxiv updated 2013/01/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We consider a partially observable Markov decision problem (POMDP) that models a class of sequencing problems. Although POMDPs are typically intractable, our formulation admits tractable solution. Instead of maintaining a value function over a high-dimensional set of belief states, we reduce the state space to one of smaller dimension, in which grid-based dynamic programming techniques are effective. We develop an error bound for the resulting approximation, and discuss an application of the model to a problem in targeted advertising.