Manifold constrained finite gaussian mixtures

Authors:
Cédric Archambeau;Michel Verleysen
Affiliations:
Machine Learning Group, Université catholique de Louvain, Louvain-la-Neuve, Belgium;Machine Learning Group, Université catholique de Louvain, Louvain-la-Neuve, Belgium
Venue:
IWANN'05 Proceedings of the 8th international conference on Artificial Neural Networks: computational Intelligence and Bioinspired Systems
Year:
2005

Citing 1
Cited 6

On convergence properties of the em algorithm for gaussian mixtures

Neural Computation

Unfolding the Manifold in Generative Topographic Mapping

HAIS '08 Proceedings of the 3rd international workshop on Hybrid Artificial Intelligence Systems
Geodesic Generative Topographic Mapping

IBERAMIA '08 Proceedings of the 11th Ibero-American conference on AI: Advances in Artificial Intelligence
On the Improvement of the Mapping Trustworthiness and Continuity of a Manifold Learning Model

IDEAL '08 Proceedings of the 9th International Conference on Intelligent Data Engineering and Automated Learning
Comparative Evaluation of Semi-supervised Geodesic GTM

HAIS '09 Proceedings of the 4th International Conference on Hybrid Artificial Intelligence Systems
Semi-supervised geodesic Generative Topographic Mapping

Pattern Recognition Letters
Semi-supervised outcome prediction for a type of human brain tumour using partially labeled MRS information

IDEAL'09 Proceedings of the 10th international conference on Intelligent data engineering and automated learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

In many practical applications, the data is organized along a manifold of lower dimension than the dimension of the embedding space. This additional information can be used when learning the model parameters of Gaussian mixtures. Based on a mismatch measure between the Euclidian and the geodesic distance, manifold constrained responsibilities are introduced. Experiments in density estimation show that manifold Gaussian mixtures outperform ordinary Gaussian mixtures.