Bayesian Wasserstein Repulsive Gaussian Mixture Models
By: Weipeng Huang, Tin Lok James Ng
Potential Business Impact:
Finds groups of things that are clearly different.
We develop the Bayesian Wasserstein repulsive Gaussian mixture model that promotes well-separated clusters. Unlike existing repulsive mixture approaches that focus on separating the component means, our method encourages separation between mixture components based on the Wasserstein distance. We establish posterior contraction rates within the framework of nonparametric density estimation. Posterior sampling is performed using a blocked-collapsed Gibbs sampler. Through simulation studies and real data applications, we demonstrate the effectiveness of the proposed model.
Similar Papers
Repulsive Mixture Model with Projection Determinantal Point Process
Methodology
Finds clear groups in messy data.
Slicing the Gaussian Mixture Wasserstein Distance
Machine Learning (CS)
Makes computer models of data faster and better.
Robust and Scalable Variational Bayes
Machine Learning (Stat)
Cleans messy data for better computer learning.