| ID:
596837.0,
MPI für Intelligente Systeme (ehemals Max-Planck-Institut für Metallforschung) / Abt. Schölkopf (Empirical Inference) |
| Probabilistic latent variable models for distinguishing between cause and effect |
| Authors: | Mooij, J. M.; Stegle, O.; Janzing, D.; Zhang, K.; Schölkopf, B. |
| Publisher: | Curran |
| Place of Publication: | Red Hook, NY |
| Date of Publication (YYYY-MM-DD): | 2011-06-01 |
| Title of Proceedings: | 24th Annual Conference on Neural Information Processing Systems (NIPS 2010) |
| Start Page: | 1687 |
| End Page: | 1695 |
| Title of Series: | Advances in neural information processing systems |
| Volume (in Series): | 23 |
| Physical Description: | 8 |
| Name of Conference/Meeting: | 24th Annual Conference on Neural Information Processing Systems (NIPS 2010) |
| Place of Conference/Meeting: | Vancouver, BC, Canada |
(Start) Date of Conference/Meeting
(YYYY-MM-DD): | 2010-12-06 |
End Date of Conference/Meeting
(YYYY-MM-DD): | 2010-12-09 |
| Review Status: | not specified |
| Audience: | Not Specified |
| Intended Educational Use: | No |
| Abstract / Description: | We propose a novel method for inferring whether X causes Y or vice versa from joint observations of X and Y. The basic idea is to model the observed data using probabilistic
latent variable models, which incorporate the effects of unobserved noise. To this end, we consider the hypothetical effect variable to be a function of the hypothetical cause variable and an independent noise term (not necessarily additive). An important novel aspect of our work is that we do not restrict the model class, but instead put general non-parametric priors on this function and on the distribution of the cause. The causal direction can then
be inferred by using standard Bayesian model selection. We evaluate our approach on synthetic data and real-world data and report encouraging results. |
| External Publication Status: | published |
| Document Type: | Conference-Paper |
| Communicated by: | Heide Klooz |
| Affiliations: | MPI für Intelligente Systeme/Abt. Schölkopf
|
| Identifiers: | URL:http://www.kyb.tuebingen.mpg.de//fileadmin/user_up...
LOCALID:6767 |
|
Advanced
