Abstract
We propose an Infinite Latent Generalized Linear Model (ILGLM), a Dirichlet process mixture of generalized linear model in latent space for classification problem. In ILGLM, we assume latent variable z n is generated from a low-dimensional DPM model in latent space, and the corresponding observed feature x n and class label y n are generated from some latent probability model and local linear classification model independently conditioned on z n . Then in ILGLM, we will jointly learn the latent variable model and multiple local generalized linear model under the framework of Dirichlet process mixture. On one hand, ILGLM can model the multiple local linearity of data distribution adaptively according to data complexity; on the other hand, it avoid the curse of dimensionality problem. ILGLM can be extended to semi-supervised setting, training the model using both labeled and unlabeled data. Because ILGLM is a general model framework, it can incorporate any kind of latent variable models and linear classification models. Then we implement ILGLM based on Factor Analysis and MultiNomial Logit model, which results in the Infinite Latent MultiNomial Logit (ILMNL) model as an example of ILGLM. We also develop an approximate posterior inference algorithm for ILMNL using Gibbs sampling. Experiments on several real-world datasets demonstrate the advantages of ILMNL in dealing with high-dimensional data classification problems compared with competitive models.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Andrieu, C., de Freitas, N., Doucet, A., Jordan, M.I.: An introduction to MCMC for machine learning. Machine Learning. Machine Learning 50(1-2), 5–43 (2003)
Antoniak, C.E.: Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems. Annals of Statistics 2, 1152–1174 (1974)
Yu, S., Yu, K., Tresp, V., Kriegel, H.P., Wu, M.: Supervised probabilistic principal component analysis. In: KDD, Philadelphia, PA, USA, pp. 464–473. ACM (August 2006)
Rasmussen, C.E., Ghahramani, Z.: Infinite mixtures of Gaussian process experts. In: Vancouver, British Columbia, Canada, Vancouver, British Columbia, Canada, pp. 881–888. MIT Press (December 2001)
Shahbaba, B., Neal, R.M.: Nonlinear models using Dirichlet process mixtures. Journal of Machine Learning Research 10, 1829–1850 (2009)
Hannah, L., Blei, D.M., Powell, W.B.: Dirichlet process mixtures of generalized linear models. Journal of Machine Learning Research 12, 1923–1953 (2011)
Zhu, J., Chen, N., Xing, E.P.: Infinite svm: a Dirichlet process mixture of large-margin kernel machines. In: ICML, Bellevue, Washington, USA, June 2011, pp. 617–624. Omnipress (2011)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer (2006)
Rish, I., Grabarnik, G., Cecchi, G.A., Pereira, F., Gordon, G.J.: Closed-form supervised dimensionality reduction with generalized linear models. In: ICML, Helsinki, Finland, pp. 832–839. ACM (June 2008)
Neal, R.M.: Markov chain sampling methods for Dirichlet process mixture models. Journal of Computational and Graphical Statistics 9(2), 249–265 (2000)
Blake, C.L., Merz, C.J.: UCI repository of machine learning databases (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Luo, J., Ren, J. (2014). An Infinite Latent Generalized Linear Model. In: Li, F., Li, G., Hwang, Sw., Yao, B., Zhang, Z. (eds) Web-Age Information Management. WAIM 2014. Lecture Notes in Computer Science, vol 8485. Springer, Cham. https://doi.org/10.1007/978-3-319-08010-9_18
Download citation
DOI: https://doi.org/10.1007/978-3-319-08010-9_18
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08009-3
Online ISBN: 978-3-319-08010-9
eBook Packages: Computer ScienceComputer Science (R0)