Timezone: »

Non-parametric Approximate Dynamic Programming via the Kernel Method
Nikhil Bhat · Ciamac C Moallemi · Vivek Farias

Thu Dec 06 02:00 PM -- 12:00 AM (PST) @ Harrah’s Special Events Center 2nd Floor

This paper presents a novel non-parametric approximate dynamic programming (ADP) algorithm that enjoys graceful, dimension-independent 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 non-parametric procedure is competitive with parametric ADP approaches.

Author Information

Nikhil Bhat (Columbia University)
Ciamac C Moallemi (Columbia University)
Vivek Farias (Massachusetts Institute of Technology)

More from the Same Authors