Online Mean Field Approximation for Automated Experimentation

Shaona Ghosh, Adam Prügel-Bennett
Proceedings of The 4th Workshop on Machine Learning for Interactive Systems at ICML 2015, PMLR 43:31-35, 2015.

Abstract

In this paper, we propose a semi-supervised online graph labelling method that affords early learning capability. We use mean field approximation for predicting the unknown labels of the vertices of the graph with high accuracy on the standard benchmark datasets. The minimum cut is the energy function of our probabilistic model that encodes the uncertainty about the labels of the vertices. Our method shows that it can learn early given any choice of experiments that may take place in the automated experimentation systems used for scientific discovery.

Cite this Paper

BibTeX
@InProceedings{pmlr-v43-ghosh15, title = {Online Mean Field Approximation for Automated Experimentation}, author = {Ghosh, Shaona and Prügel-Bennett, Adam}, booktitle = {Proceedings of The 4th Workshop on Machine Learning for Interactive Systems at ICML 2015}, pages = {31--35}, year = {2015}, editor = {Cuayáhuitl, Heriberto and Dethlefs, Nina and Frommberger, Lutz and Van Otterlo, Martijn and Pietquin, Olivier}, volume = {43}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {11 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v43/ghosh15.pdf}, url = {https://proceedings.mlr.press/v43/ghosh15.html}, abstract = {In this paper, we propose a semi-supervised online graph labelling method that affords early learning capability. We use mean field approximation for predicting the unknown labels of the vertices of the graph with high accuracy on the standard benchmark datasets. The minimum cut is the energy function of our probabilistic model that encodes the uncertainty about the labels of the vertices. Our method shows that it can learn early given any choice of experiments that may take place in the automated experimentation systems used for scientific discovery.} }
Endnote
%0 Conference Paper %T Online Mean Field Approximation for Automated Experimentation %A Shaona Ghosh %A Adam Prügel-Bennett %B Proceedings of The 4th Workshop on Machine Learning for Interactive Systems at ICML 2015 %C Proceedings of Machine Learning Research %D 2015 %E Heriberto Cuayáhuitl %E Nina Dethlefs %E Lutz Frommberger %E Martijn Van Otterlo %E Olivier Pietquin %F pmlr-v43-ghosh15 %I PMLR %P 31--35 %U https://proceedings.mlr.press/v43/ghosh15.html %V 43 %X In this paper, we propose a semi-supervised online graph labelling method that affords early learning capability. We use mean field approximation for predicting the unknown labels of the vertices of the graph with high accuracy on the standard benchmark datasets. The minimum cut is the energy function of our probabilistic model that encodes the uncertainty about the labels of the vertices. Our method shows that it can learn early given any choice of experiments that may take place in the automated experimentation systems used for scientific discovery.
RIS
TY - CPAPER TI - Online Mean Field Approximation for Automated Experimentation AU - Shaona Ghosh AU - Adam Prügel-Bennett BT - Proceedings of The 4th Workshop on Machine Learning for Interactive Systems at ICML 2015 DA - 2015/06/18 ED - Heriberto Cuayáhuitl ED - Nina Dethlefs ED - Lutz Frommberger ED - Martijn Van Otterlo ED - Olivier Pietquin ID - pmlr-v43-ghosh15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 43 SP - 31 EP - 35 L1 - http://proceedings.mlr.press/v43/ghosh15.pdf UR - https://proceedings.mlr.press/v43/ghosh15.html AB - In this paper, we propose a semi-supervised online graph labelling method that affords early learning capability. We use mean field approximation for predicting the unknown labels of the vertices of the graph with high accuracy on the standard benchmark datasets. The minimum cut is the energy function of our probabilistic model that encodes the uncertainty about the labels of the vertices. Our method shows that it can learn early given any choice of experiments that may take place in the automated experimentation systems used for scientific discovery. ER -
APA
Ghosh, S. & Prügel-Bennett, A.. (2015). Online Mean Field Approximation for Automated Experimentation. Proceedings of The 4th Workshop on Machine Learning for Interactive Systems at ICML 2015, in Proceedings of Machine Learning Research 43:31-35 Available from https://proceedings.mlr.press/v43/ghosh15.html.

Related Material