|
|
|
|
| LEADER |
01294 am a22001933u 4500 |
| 001 |
89425 |
| 042 |
|
|
|a dc
|
| 100 |
1 |
0 |
|a Bhat, Nikhil
|e author
|
| 100 |
1 |
0 |
|a Sloan School of Management
|e contributor
|
| 100 |
1 |
0 |
|a Farias, Vivek F.
|e contributor
|
| 700 |
1 |
0 |
|a Farias, Vivek F.
|e author
|
| 700 |
1 |
0 |
|a Moallemi, Ciamac C.
|e author
|
| 245 |
0 |
0 |
|a Non-Parametric Approximate Dynamic Programming via the Kernel Method
|
| 260 |
|
|
|b Neural Information Processing Systems Foundation,
|c 2014-09-11T12:56:58Z.
|
| 856 |
|
|
|z Get fulltext
|u http://hdl.handle.net/1721.1/89425
|
| 520 |
|
|
|a 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.
|
| 546 |
|
|
|a en_US
|
| 655 |
7 |
|
|a Article
|
| 773 |
|
|
|t Proceedings of the 2012 Neural Information Processing Systems Conference
|
