|
|
| Record Number | 10266 |
| Reference Type | Conference Proceedings |
| Author(s) | Ting, J.;Kalakrishnan, M.;Vijayakumar, S.;Schaal, S. |
| Year | 2008 |
| Title | Bayesian kernel shaping for learning control |
| Journal/Conference/Book Title | Advances in Neural Information Processing Systems 21 (NIPS 2008) |
| Keywords | locally weighted learning, kernel regression, bayesian learning, nonstationary processes |
Abstract |
In kernel-based regression learning, optimizing each kernel individually is useful
when the data density, curvature of regression surfaces (or decision boundaries)
or magnitude of output noise varies spatially. Previous work has suggested gradi-
ent descent techniques or complex statistical hypothesis methods for local kernel
shaping, typically requiring some amount of manual tuning of meta parameters.
We introduce a Bayesian formulation of nonparametric regression that, with the
help of variational approximations, results in an EM-like algorithm for simulta-
neous estimation of regression and kernel parameters. The algorithm is computa-
tionally efficient, requires no sampling, automatically rejects outliers and has only
one prior to be specified. It can be used for nonparametric regression with local
polynomials or as a novel method to achieve nonstationary regression with Gaus-
sian processes. Our methods are particularly useful for learning control, where
reliable estimation of local tangent planes is essential for adaptive controllers and
reinforcement learning. We evaluate our methods on several synthetic data sets
and on an actual robot which learns a task-level control law.
|
| Notes | clmc |
| URL(s) | http://www-clmc.usc.edu/publications/T/ting-NIPS2008.pdf
|
| Editor(s) | Koller, D.;Bengio, Y.;Schuurmans, D.;Bottou, L.;Culotta, A. |
| Place Published | Vancouver, BC, Dec. 7-10 |
| Publisher | Cambridge, MA: MIT Press |
| Short Title | Bayesian kernel shaping for learning control |
| Papers are available as Adobe PDF ".pdf" files. Adobe Reader is available for free for all computer platforms.
|
|
|
|
|
Page last modified on August 10, 2006, at 06:47 PM
|
|
