Generative Kernels for Exponential Families

Arvind Agarwal, Hal Daumé III
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:85-92, 2011.

Abstract

In this paper, we propose a family of kernels for the data distributions belonging to the exponential family. We call these kernels generative kernels because they take into account the generative process of the data. Our proposed method considers the geometry of the data distribution to build a set of efficient closed-form kernels best suited for that distribution. We compare our generative kernels on multinomial data and observe improved empirical performance across the board. Moreover, our generative kernels perform significantly better when training size is small, an important property of the generative models.

Cite this Paper


BibTeX
@InProceedings{pmlr-v15-agarwal11b, title = {Generative Kernels for Exponential Families}, author = {Agarwal, Arvind and Daum\'e, III, Hal}, booktitle = {Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics}, pages = {85--92}, year = {2011}, editor = {Gordon, Geoffrey and Dunson, David and Dudík, Miroslav}, volume = {15}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {11--13 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v15/agarwal11b/agarwal11b.pdf}, url = {https://proceedings.mlr.press/v15/agarwal11b.html}, abstract = {In this paper, we propose a family of kernels for the data distributions belonging to the exponential family. We call these kernels generative kernels because they take into account the generative process of the data. Our proposed method considers the geometry of the data distribution to build a set of efficient closed-form kernels best suited for that distribution. We compare our generative kernels on multinomial data and observe improved empirical performance across the board. Moreover, our generative kernels perform significantly better when training size is small, an important property of the generative models.} }
Endnote
%0 Conference Paper %T Generative Kernels for Exponential Families %A Arvind Agarwal %A Hal Daumé, III %B Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2011 %E Geoffrey Gordon %E David Dunson %E Miroslav Dudík %F pmlr-v15-agarwal11b %I PMLR %P 85--92 %U https://proceedings.mlr.press/v15/agarwal11b.html %V 15 %X In this paper, we propose a family of kernels for the data distributions belonging to the exponential family. We call these kernels generative kernels because they take into account the generative process of the data. Our proposed method considers the geometry of the data distribution to build a set of efficient closed-form kernels best suited for that distribution. We compare our generative kernels on multinomial data and observe improved empirical performance across the board. Moreover, our generative kernels perform significantly better when training size is small, an important property of the generative models.
RIS
TY - CPAPER TI - Generative Kernels for Exponential Families AU - Arvind Agarwal AU - Hal Daumé, III BT - Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics DA - 2011/06/14 ED - Geoffrey Gordon ED - David Dunson ED - Miroslav Dudík ID - pmlr-v15-agarwal11b PB - PMLR DP - Proceedings of Machine Learning Research VL - 15 SP - 85 EP - 92 L1 - http://proceedings.mlr.press/v15/agarwal11b/agarwal11b.pdf UR - https://proceedings.mlr.press/v15/agarwal11b.html AB - In this paper, we propose a family of kernels for the data distributions belonging to the exponential family. We call these kernels generative kernels because they take into account the generative process of the data. Our proposed method considers the geometry of the data distribution to build a set of efficient closed-form kernels best suited for that distribution. We compare our generative kernels on multinomial data and observe improved empirical performance across the board. Moreover, our generative kernels perform significantly better when training size is small, an important property of the generative models. ER -
APA
Agarwal, A. & Daumé, III, H.. (2011). Generative Kernels for Exponential Families. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 15:85-92 Available from https://proceedings.mlr.press/v15/agarwal11b.html.

Related Material