Anchored Discrete Factor Analysis. Halpern, Y., Horng, S., & Sontag, D. In arXiv:1511.03299, 2015.
Paper abstract bibtex We present a semi-supervised learning algorithm for learning discrete factor analysis models with arbitrary structure on the latent variables. Our algorithm assumes that every latent variable has an "anchor", an observed variable with only that latent variable as its parent. Given such anchors, we show that it is possible to consistently recover moments of the latent variables and use these moments to learn complete models. We also introduce a new technique for improving the robustness of method-of-moment algorithms by optimizing over the marginal polytope or its relaxations. We evaluate our algorithm using two real-world tasks, tag prediction on questions from the Stack Overflow website and medical diagnosis in an emergency department.
@inproceedings{HalpernEtAl_arxiv15,
author = {Yoni Halpern and Steven Horng and David Sontag},
title = {Anchored Discrete Factor Analysis},
booktitle = {arXiv:1511.03299},
year = {2015},
keywords = {Machine learning, Unsupervised learning, Health care},
url_Paper = {http://arxiv.org/pdf/1511.03299.pdf},
abstract = {We present a semi-supervised learning algorithm for learning discrete factor analysis models with arbitrary structure on the latent variables. Our algorithm assumes that every latent variable has an "anchor", an observed variable with only that latent variable as its parent. Given such anchors, we show that it is possible to consistently recover moments of the latent variables and use these moments to learn complete models. We also introduce a new technique for improving the robustness of method-of-moment algorithms by optimizing over the marginal polytope or its relaxations. We evaluate our algorithm using two real-world tasks, tag prediction on questions from the Stack Overflow website and medical diagnosis in an emergency department.}
}
Downloads: 0
{"_id":"so6aNceT8JPJRYY8s","bibbaseid":"halpern-horng-sontag-anchoreddiscretefactoranalysis-2015","author_short":["Halpern, Y.","Horng, S.","Sontag, D."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Yoni"],"propositions":[],"lastnames":["Halpern"],"suffixes":[]},{"firstnames":["Steven"],"propositions":[],"lastnames":["Horng"],"suffixes":[]},{"firstnames":["David"],"propositions":[],"lastnames":["Sontag"],"suffixes":[]}],"title":"Anchored Discrete Factor Analysis","booktitle":"arXiv:1511.03299","year":"2015","keywords":"Machine learning, Unsupervised learning, Health care","url_paper":"http://arxiv.org/pdf/1511.03299.pdf","abstract":"We present a semi-supervised learning algorithm for learning discrete factor analysis models with arbitrary structure on the latent variables. Our algorithm assumes that every latent variable has an \"anchor\", an observed variable with only that latent variable as its parent. Given such anchors, we show that it is possible to consistently recover moments of the latent variables and use these moments to learn complete models. We also introduce a new technique for improving the robustness of method-of-moment algorithms by optimizing over the marginal polytope or its relaxations. We evaluate our algorithm using two real-world tasks, tag prediction on questions from the Stack Overflow website and medical diagnosis in an emergency department.","bibtex":"@inproceedings{HalpernEtAl_arxiv15,\n author = {Yoni Halpern and Steven Horng and David Sontag},\n title = {Anchored Discrete Factor Analysis},\n booktitle = {arXiv:1511.03299},\n year = {2015},\n keywords = {Machine learning, Unsupervised learning, Health care},\n url_Paper = {http://arxiv.org/pdf/1511.03299.pdf},\n abstract = {We present a semi-supervised learning algorithm for learning discrete factor analysis models with arbitrary structure on the latent variables. Our algorithm assumes that every latent variable has an \"anchor\", an observed variable with only that latent variable as its parent. Given such anchors, we show that it is possible to consistently recover moments of the latent variables and use these moments to learn complete models. We also introduce a new technique for improving the robustness of method-of-moment algorithms by optimizing over the marginal polytope or its relaxations. We evaluate our algorithm using two real-world tasks, tag prediction on questions from the Stack Overflow website and medical diagnosis in an emergency department.}\n}\n\n","author_short":["Halpern, Y.","Horng, S.","Sontag, D."],"key":"HalpernEtAl_arxiv15","id":"HalpernEtAl_arxiv15","bibbaseid":"halpern-horng-sontag-anchoreddiscretefactoranalysis-2015","role":"author","urls":{" paper":"http://arxiv.org/pdf/1511.03299.pdf"},"keyword":["Machine learning","Unsupervised learning","Health care"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"bibtype":"inproceedings","biburl":"http://people.csail.mit.edu/dsontag/papers/bibtex/david_sontag_papers_all.bib","dataSources":["g3ofqhxNQWsRWkCrp"],"keywords":["machine learning","unsupervised learning","health care"],"search_terms":["anchored","discrete","factor","analysis","halpern","horng","sontag"],"title":"Anchored Discrete Factor Analysis","year":2015}