TY - GEN
T1 - Discovery of Optimal Solution Horizons in Non-Stationary Markov Decision Processes with Unbounded Rewards
AU - Neustroev, Greg
AU - de Weerdt, Mathijs
AU - Verzijlbergh, Remco
N1 - Conference code: 29
PY - 2019
Y1 - 2019
N2 - Infinite-horizon non-stationary Markov decision processes provide a general framework to model many real-life decision-making problems, e.g., planning equipment maintenance. Unfortunately, these problems are notoriously difficult to solve, due to their infinite dimensionality. Often, only the optimality of the initial action is of importance to the decision-maker: once it has been identified, the procedure can be repeated to generate a plan of arbitrary length. The optimal initial action can be identified by finding a time horizon so long that data beyond it has no effect on the initial decision. This horizon is known as a solution horizon and can be discovered by considering a series of truncations of the problem until a stopping rule guaranteeing initial decision optimality is satisfied. We present such a stopping rule for problems with unbounded rewards. Given a candidate policy, the rule uses a mathematical program that searches for other possibly optimal initial actions within the space of feasible truncations. If no better action can be found, the candidate action is deemed optimal. Our rule runs faster than comparable rules and discovers shorter, more efficient solution horizons.
AB - Infinite-horizon non-stationary Markov decision processes provide a general framework to model many real-life decision-making problems, e.g., planning equipment maintenance. Unfortunately, these problems are notoriously difficult to solve, due to their infinite dimensionality. Often, only the optimality of the initial action is of importance to the decision-maker: once it has been identified, the procedure can be repeated to generate a plan of arbitrary length. The optimal initial action can be identified by finding a time horizon so long that data beyond it has no effect on the initial decision. This horizon is known as a solution horizon and can be discovered by considering a series of truncations of the problem until a stopping rule guaranteeing initial decision optimality is satisfied. We present such a stopping rule for problems with unbounded rewards. Given a candidate policy, the rule uses a mathematical program that searches for other possibly optimal initial actions within the space of feasible truncations. If no better action can be found, the candidate action is deemed optimal. Our rule runs faster than comparable rules and discovers shorter, more efficient solution horizons.
UR - http://www.scopus.com/inward/record.url?scp=85085621868&partnerID=8YFLogxK
M3 - Conference contribution
VL - 29
T3 - Proceedings International Conference on Automated Planning and Scheduling, ICAPS
SP - 292
EP - 300
BT - Proceedings of the 29th International Conference on Automated Planning and Scheduling, ICAPS 2019
A2 - Benton, J.
A2 - Lipovetzky, Nir
A2 - Onaindia, Eva
A2 - Smith, David E.
A2 - Srivastava, Siddharth
PB - Association for the Advancement of Artificial Intelligence (AAAI)
T2 - 29th International Conference on Automated Planning and Scheduling
Y2 - 11 July 2019 through 15 July 2019
ER -