Learning probability by comparison

Taisuke Sato
; Proceedings of The 3rd International Workshop on Advanced Methodologies for Bayesian Networks, PMLR 73:5-5, 2017.

Abstract

Learning probability by probabilistic modeling is a major task in statistical machine learning and it has traditionally been supported by maximum likelihood estimation applied to generative models or by a local maximizer applied to discriminative models. In this talk, we introduce a third approach, an innovative one that learns probability by comparing probabilistic events. In our approach, we give the ranking of probabilistic events and the system learns a probability distribution so that the ranking is well respected. We implemented this approach in PRISM, a logic-based probabilistic programming language, and conducted learning experiments with real data for models described by PRISM programs.

Cite this Paper

BibTeX
@InProceedings{pmlr-v73-sato17a, title = {Learning probability by comparison}, author = {Taisuke Sato}, booktitle = {Proceedings of The 3rd International Workshop on Advanced Methodologies for Bayesian Networks}, pages = {5--5}, year = {2017}, editor = {Antti Hyttinen and Joe Suzuki and Brandon Malone}, volume = {73}, series = {Proceedings of Machine Learning Research}, month = {20--22 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v73/sato17a/sato17a.pdf}, url = {http://proceedings.mlr.press/v73/sato17a.html}, abstract = {Learning probability by probabilistic modeling is a major task in statistical machine learning and it has traditionally been supported by maximum likelihood estimation applied to generative models or by a local maximizer applied to discriminative models. In this talk, we introduce a third approach, an innovative one that learns probability by comparing probabilistic events. In our approach, we give the ranking of probabilistic events and the system learns a probability distribution so that the ranking is well respected. We implemented this approach in PRISM, a logic-based probabilistic programming language, and conducted learning experiments with real data for models described by PRISM programs.} }
Endnote
%0 Conference Paper %T Learning probability by comparison %A Taisuke Sato %B Proceedings of The 3rd International Workshop on Advanced Methodologies for Bayesian Networks %C Proceedings of Machine Learning Research %D 2017 %E Antti Hyttinen %E Joe Suzuki %E Brandon Malone %F pmlr-v73-sato17a %I PMLR %J Proceedings of Machine Learning Research %P 5--5 %U http://proceedings.mlr.press %V 73 %W PMLR %X Learning probability by probabilistic modeling is a major task in statistical machine learning and it has traditionally been supported by maximum likelihood estimation applied to generative models or by a local maximizer applied to discriminative models. In this talk, we introduce a third approach, an innovative one that learns probability by comparing probabilistic events. In our approach, we give the ranking of probabilistic events and the system learns a probability distribution so that the ranking is well respected. We implemented this approach in PRISM, a logic-based probabilistic programming language, and conducted learning experiments with real data for models described by PRISM programs.
APA
Sato, T.. (2017). Learning probability by comparison. Proceedings of The 3rd International Workshop on Advanced Methodologies for Bayesian Networks, in PMLR 73:5-5

Related Material