Infinite Markov-Switching Maximum Entropy Discrimination Machines

Sotirios Chatzis
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):729-737, 2013.

Abstract

In this paper, we present a method that combines the merits of Bayesian nonparametrics, specifically stick-breaking priors, and large-margin kernel machines in the context of sequential data classification. The proposed model postulates a set of (theoretically) infinite interdependent large-margin classifiers as model components, that robustly capture local nonlinearity of complex data. The postulated large-margin classifiers are connected in the context of a Markov-switching construction that allows for capturing complex temporal dynamics in the modeled datasets. Appropriate stick-breaking priors are imposed over the component switching mechanism of our model to allow for data-driven determination of the optimal number of component large-margin classifiers, under a standard nonparametric Bayesian inference scheme. Efficient model training is performed under the maximum entropy discrimination (MED) framework, which integrates the large-margin principle with Bayesian posterior inference. We evaluate our method using several real-world datasets, and compare it to state-of-the-art alternatives.

Cite this Paper

BibTeX
@InProceedings{pmlr-v28-chatzis13, title = {Infinite Markov-Switching Maximum Entropy Discrimination Machines}, author = {Chatzis, Sotirios}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {729--737}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/chatzis13.pdf}, url = {https://proceedings.mlr.press/v28/chatzis13.html}, abstract = {In this paper, we present a method that combines the merits of Bayesian nonparametrics, specifically stick-breaking priors, and large-margin kernel machines in the context of sequential data classification. The proposed model postulates a set of (theoretically) infinite interdependent large-margin classifiers as model components, that robustly capture local nonlinearity of complex data. The postulated large-margin classifiers are connected in the context of a Markov-switching construction that allows for capturing complex temporal dynamics in the modeled datasets. Appropriate stick-breaking priors are imposed over the component switching mechanism of our model to allow for data-driven determination of the optimal number of component large-margin classifiers, under a standard nonparametric Bayesian inference scheme. Efficient model training is performed under the maximum entropy discrimination (MED) framework, which integrates the large-margin principle with Bayesian posterior inference. We evaluate our method using several real-world datasets, and compare it to state-of-the-art alternatives. } }
Endnote
%0 Conference Paper %T Infinite Markov-Switching Maximum Entropy Discrimination Machines %A Sotirios Chatzis %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-chatzis13 %I PMLR %P 729--737 %U https://proceedings.mlr.press/v28/chatzis13.html %V 28 %N 3 %X In this paper, we present a method that combines the merits of Bayesian nonparametrics, specifically stick-breaking priors, and large-margin kernel machines in the context of sequential data classification. The proposed model postulates a set of (theoretically) infinite interdependent large-margin classifiers as model components, that robustly capture local nonlinearity of complex data. The postulated large-margin classifiers are connected in the context of a Markov-switching construction that allows for capturing complex temporal dynamics in the modeled datasets. Appropriate stick-breaking priors are imposed over the component switching mechanism of our model to allow for data-driven determination of the optimal number of component large-margin classifiers, under a standard nonparametric Bayesian inference scheme. Efficient model training is performed under the maximum entropy discrimination (MED) framework, which integrates the large-margin principle with Bayesian posterior inference. We evaluate our method using several real-world datasets, and compare it to state-of-the-art alternatives.
RIS
TY - CPAPER TI - Infinite Markov-Switching Maximum Entropy Discrimination Machines AU - Sotirios Chatzis BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/26 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-chatzis13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 3 SP - 729 EP - 737 L1 - http://proceedings.mlr.press/v28/chatzis13.pdf UR - https://proceedings.mlr.press/v28/chatzis13.html AB - In this paper, we present a method that combines the merits of Bayesian nonparametrics, specifically stick-breaking priors, and large-margin kernel machines in the context of sequential data classification. The proposed model postulates a set of (theoretically) infinite interdependent large-margin classifiers as model components, that robustly capture local nonlinearity of complex data. The postulated large-margin classifiers are connected in the context of a Markov-switching construction that allows for capturing complex temporal dynamics in the modeled datasets. Appropriate stick-breaking priors are imposed over the component switching mechanism of our model to allow for data-driven determination of the optimal number of component large-margin classifiers, under a standard nonparametric Bayesian inference scheme. Efficient model training is performed under the maximum entropy discrimination (MED) framework, which integrates the large-margin principle with Bayesian posterior inference. We evaluate our method using several real-world datasets, and compare it to state-of-the-art alternatives. ER -
APA
Chatzis, S.. (2013). Infinite Markov-Switching Maximum Entropy Discrimination Machines. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(3):729-737 Available from https://proceedings.mlr.press/v28/chatzis13.html.

Related Material