Constrained Bayesian inference for low rank multitask learning

Research output: Contribution to conferencePaperpeer-review

Abstract

We present a novel approach for constrained Bayesian inference. Unlike current methods, our approach does not require convexity of the constraint set. We reduce the constrained variational inference to a parametric optimization over the feasible set of densities and propose a general recipe for such problems. We apply the proposed constrained Bayesian inference approach to multitask learning subject to rank constraints on the weight matrix. Further, constrained parameter estimation is applied to recover the sparse conditional independence structure encoded by prior precision matrices. Our approach is motivated by reverse inference for high dimensional functional neuroimaging, a domain where the high dimensionality and small number of examples requires the use of constraints to ensure meaningful and effective models. For this application, we propose a model that jointly learns a weight matrix and the prior inverse covariance structure between different tasks. We present experimental validation showing that the proposed approach outperforms strong baseline models in terms of predictive performance and structure recovery.

Original languageEnglish (US)
Pages341-350
Number of pages10
StatePublished - Nov 28 2013
Externally publishedYes
Event29th Conference on Uncertainty in Artificial Intelligence, UAI 2013 - Bellevue, WA, United States
Duration: Jul 11 2013Jul 15 2013

Other

Other29th Conference on Uncertainty in Artificial Intelligence, UAI 2013
Country/TerritoryUnited States
CityBellevue, WA
Period7/11/137/15/13

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Constrained Bayesian inference for low rank multitask learning'. Together they form a unique fingerprint.

Cite this