Skip to yearly menu bar Skip to main content


DP-GP-LVM: A Bayesian Non-Parametric Model for Learning Multivariate Dependency Structures

Andrew R Lawrence · Carl Henrik Ek · Neill Campbell

Pacific Ballroom #221

Keywords: [ Unsupervised Learning ] [ Interpretability ] [ Generative Models ] [ Gaussian Processes ] [ Bayesian Nonparametrics ]


We present a non-parametric Bayesian latent variable model capable of learning dependency structures across dimensions in a multivariate setting. Our approach is based on flexible Gaussian process priors for the generative mappings and interchangeable Dirichlet process priors to learn the structure. The introduction of the Dirichlet process as a specific structural prior allows our model to circumvent issues associated with previous Gaussian process latent variable models. Inference is performed by deriving an efficient variational bound on the marginal log-likelihood of the model. We demonstrate the efficacy of our approach via analysis of discovered structure and superior quantitative performance on missing data imputation.

Live content is unavailable. Log in and register to view live content