Statistics > Machine Learning
[Submitted on 2 May 2013 (v1), revised 3 May 2013 (this version, v2), latest version 20 Jan 2015 (v6)]
Title:Learning Mixtures of Bernoulli Templates by Two-Round EM with Performance Guarantee
View PDFAbstract:Dasgupta showed that a two-round variant of the EM algorithm can learn mixture of Gaussian distributions with near optimal precision with high probability if the Gaussian distributions are well separated and if the dimension is sufficiently high. In this paper, we generalize their theory to learning mixture of high-dimensional Bernoulli templates. Each template is a binary vector, and a template generates examples by randomly switching its binary components independently with a certain probability. In computer vision applications, a binary vector is a feature map of an image, where each binary component indicates whether a feature or structure is present or absent within a certain cell of the image domain. A Bernoulli template can be considered a statistical model for images of objects (or parts of objects) from the same category. We show that the two-round EM algorithm can learn mixture of Bernoulli templates with near optimal precision with high probability, if the Bernoulli templates are sufficiently different and if the number of features is sufficiently high. We illustrate the theoretical results by synthetic and real examples.
Submission history
From: Adrian Barbu [view email][v1] Thu, 2 May 2013 00:31:45 UTC (1,050 KB)
[v2] Fri, 3 May 2013 09:55:22 UTC (1,050 KB)
[v3] Thu, 5 Sep 2013 16:46:21 UTC (1,327 KB)
[v4] Tue, 10 Dec 2013 16:44:37 UTC (1,327 KB)
[v5] Wed, 13 Aug 2014 21:23:36 UTC (1,444 KB)
[v6] Tue, 20 Jan 2015 03:16:21 UTC (1,419 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.