TY - GEN
T1 - Planning for welfare to work
AU - Liangrong, Yi
AU - Finkel, Raphael
AU - Goldsmith, Judy
PY - 2008
Y1 - 2008
N2 - We are interested in building decision-support software for social welfare case managers. Our model in the form of a factored Markov decision process is so complex that a standard factored MDP solver was unable to solve it efficiently. We discuss factors contributing to the complexity of the model, then present a receding horizon planner that offers a rough policy quickly. Our planner computes locally, both in the sense of only offering one action suggestion at a time (rather than a complete policy) and because it starts from an initial state and considers only states reachable from there in its calculations.
AB - We are interested in building decision-support software for social welfare case managers. Our model in the form of a factored Markov decision process is so complex that a standard factored MDP solver was unable to solve it efficiently. We discuss factors contributing to the complexity of the model, then present a receding horizon planner that offers a rough policy quickly. Our planner computes locally, both in the sense of only offering one action suggestion at a time (rather than a complete policy) and because it starts from an initial state and considers only states reachable from there in its calculations.
UR - http://www.scopus.com/inward/record.url?scp=55849104278&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=55849104278&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:55849104278
SN - 9781577353652
T3 - Proceedings of the 21th International Florida Artificial Intelligence Research Society Conference, FLAIRS-21
SP - 696
EP - 701
BT - Proceedings of the 21th International Florida Artificial Intelligence Research Society Conference, FLAIRS-21
T2 - 21th International Florida Artificial Intelligence Research Society Conference, FLAIRS-21
Y2 - 15 May 2008 through 17 May 2008
ER -