A conditional game for comparing approximations

Frederik Eaton
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:63-71, 2011.

Abstract

We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals.

Cite this Paper

BibTeX
@InProceedings{pmlr-v15-eaton11a, title = {A conditional game for comparing approximations}, author = {Eaton, Frederik}, booktitle = {Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics}, pages = {63--71}, 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/eaton11a/eaton11a.pdf}, url = {https://proceedings.mlr.press/v15/eaton11a.html}, abstract = {We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals. } }
Endnote
%0 Conference Paper %T A conditional game for comparing approximations %A Frederik Eaton %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-eaton11a %I PMLR %P 63--71 %U https://proceedings.mlr.press/v15/eaton11a.html %V 15 %X We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals.
RIS
TY - CPAPER TI - A conditional game for comparing approximations AU - Frederik Eaton 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-eaton11a PB - PMLR DP - Proceedings of Machine Learning Research VL - 15 SP - 63 EP - 71 L1 - http://proceedings.mlr.press/v15/eaton11a/eaton11a.pdf UR - https://proceedings.mlr.press/v15/eaton11a.html AB - We present a “conditional game” to be played between two approximate inference algorithms. We prove that exact inference is an optimal strategy and demonstrate how the game can be used to estimate the relative accuracy of two different approximations in the absence of exact marginals. ER -
APA
Eaton, F.. (2011). A conditional game for comparing approximations. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 15:63-71 Available from https://proceedings.mlr.press/v15/eaton11a.html.

Related Material