|
|
|
ID:
548528.0,
MPI für biologische Kybernetik / Biologische Kybernetik |
Nonlinear directed acyclic structure learning with weakly additive noise models |
Authors: | Tillman, R.E.; Gretton, A.; Spirtes, P. | Editors: | Bengio, Y.; Schuurmans, D.; Lafferty, J.; Williams, C.; Culotta, A. | Date of Publication (YYYY-MM-DD): | 2010-04 | Title of Proceedings: | Advances in Neural Information Processing Systems 22: 23rd Annual Conference on Neural Information Processing Systems 2009 | Start Page: | 1847 | End Page: | 1855 | Physical Description: | 9 | Audience: | Not Specified | Intended Educational Use: | No | Abstract / Description: | The recently proposed emph{additive noise model} has advantages over previous structure learning algorithms, when attempting to recover some true data generating mechanism, since it (i) does not assume linearity or Gaussianity and (ii) can recover a unique DAG rather than an equivalence class. However, its original extension to the multivariate case required enumerating all possible DAGs, and for some special distributions, e.g. linear Gaussian, the model is invertible and thus cannot be used for structure learning. We present a new approach which combines a PC style search using recent advances in kernel measures of conditional dependence with local searches for additive noise models in substructures of the equivalence class. This results in a more computationally efficient approach that is useful for arbitrary distributions even when additive noise models are invertible. Experiments with synthetic and real data show that this method is more accurate than previous methods when data are nonlinear and/or non-G aussian. | External Publication Status: | published | Document Type: | Conference-Paper |
Communicated by: | Holger Fischer | Affiliations: | MPI für biologische Kybernetik/Empirical Inference (Dept. Schölkopf)
| Identifiers: | LOCALID:6133 URL:http://nips.cc/Conferences/2009/ | |
|
|
The scope and number of records on eDoc is subject
to the collection policies defined by each institute
- see "info" button in the collection browse view.
|
|