Non-Parametric Approximate Dynamic Programming via the Kernel Method
Author(s)Bhat, Nikhil; Farias, Vivek F.; Moallemi, Ciamac C.
MetadataShow full item record
This paper presents a novel non-parametric approximate dynamic programming (ADP) algorithm that enjoys graceful approximation and sample complexity guarantees. In particular, we establish both theoretically and computationally that our proposal can serve as a viable alternative to state-of-the-art parametric ADP algorithms, freeing the designer from carefully specifying an approximation architecture. We accomplish this by developing a kernel-based mathematical program for ADP. Via a computational study on a controlled queueing network, we show that our procedure is competitive with parametric ADP approaches.
DepartmentSloan School of Management
Proceedings of the 2012 Neural Information Processing Systems Conference
Neural Information Processing Systems Foundation
Bhat, Nikhil, Vivek F. Farias, and Ciamac C. Moallemi. "Non-Parametric Approximate Dynamic Programming via the Kernel Method." The 2012 Neural Information Processing Systems Conference, Lake Tahoe, Nevada, December 3-8, 2012.
Author's final manuscript