Skip to main content

Non-Gaussian discriminative factor models via the max-margin rank-likelihood

Publication ,  Journal Article
Yuan, X; Henao, R; Tsalik, EL; Langley, RJ; Carin, L
Published in: 32nd International Conference on Machine Learning, ICML 2015
January 1, 2015

We consider the problem of discriminative factor analysis for data that are in general non-Gaussian. A Bayesian model based on the ranks of the data is proposed. We first introduce a new max-margin version of the rank-likelihood. A discriminative factor model is then developed, integrating the max-margin rank-likelihood and (linear) Bayesian support vector machines, which are also built on the max-margin principle. The discriminative factor model is further extended to the nonlinear case through mixtures of local linear classifiers, via Dirichlet processes. Fully local conjugacy of the model yields efficient inference with both Markov Chain Monte Carlo and variational Bayes approaches. Extensive experiments on benchmark and real data demonstrate superior performance of the proposed model and its potential for applications in computational biology.

Duke Scholars

Published In

32nd International Conference on Machine Learning, ICML 2015

Publication Date

January 1, 2015

Volume

2

Start / End Page

1254 / 1263
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Yuan, X., Henao, R., Tsalik, E. L., Langley, R. J., & Carin, L. (2015). Non-Gaussian discriminative factor models via the max-margin rank-likelihood. 32nd International Conference on Machine Learning, ICML 2015, 2, 1254–1263.
Yuan, X., R. Henao, E. L. Tsalik, R. J. Langley, and L. Carin. “Non-Gaussian discriminative factor models via the max-margin rank-likelihood.” 32nd International Conference on Machine Learning, ICML 2015 2 (January 1, 2015): 1254–63.
Yuan X, Henao R, Tsalik EL, Langley RJ, Carin L. Non-Gaussian discriminative factor models via the max-margin rank-likelihood. 32nd International Conference on Machine Learning, ICML 2015. 2015 Jan 1;2:1254–63.
Yuan, X., et al. “Non-Gaussian discriminative factor models via the max-margin rank-likelihood.” 32nd International Conference on Machine Learning, ICML 2015, vol. 2, Jan. 2015, pp. 1254–63.
Yuan X, Henao R, Tsalik EL, Langley RJ, Carin L. Non-Gaussian discriminative factor models via the max-margin rank-likelihood. 32nd International Conference on Machine Learning, ICML 2015. 2015 Jan 1;2:1254–1263.

Published In

32nd International Conference on Machine Learning, ICML 2015

Publication Date

January 1, 2015

Volume

2

Start / End Page

1254 / 1263