Copyright (c) 2009 All rights reserved. http://works.bepress.com/corrada_andres Recent documents in Andrés Corrada-Emmanuel en-us Sun, 31 May 2009 04:32:17 PDT 3600 http://works.bepress.com/corrada_andres/5 http://works.bepress.com/corrada_andres/5 Wed, 06 Feb 2008 10:55:51 PST Errors in map-making tasks using computer vision are sparse. We demonstrate this by considering the construction of digital elevation models that employ stereo matching algorithms to triangulate real-world points. This sparsity, coupled with a geometric theory of errors recently developed by the authors, allows for autonomous agents to calculate their own precision independently of ground truth. We connect these developments with recent advances in the mathematics of sparse signal reconstruction or compressed sensing. The theory presented here extends the autonomy of 3-D model reconstructions discovered in the 1990s to their errors. Andrés Corrada-Emmanuel Machine Learning Photogrammetry Computer Vision http://works.bepress.com/corrada_andres/4 http://works.bepress.com/corrada_andres/4 Wed, 16 Jan 2008 16:59:08 PST The precision errors in a collection of digital elevation models (DEMs) can be estimated in the presence of large but sparse correlations even when no ground truth is known. We demonstrate this by considering the problem of how to estimate the horizontal decorrelation length of DEMs produced by an automatic photogrammetric process that relies on the epipolar constraint equations. The procedure is based on a set of autonomous elevation difference equations recently proposed by us. In this paper we show that these equations can only estimate the precision errors of DEMs. The accuracy errors are unknowable since there is no ground truth. Furthermore, consideration of the invariance properties of the equations make clear that their application is limited to an imaging sensor that is accurate in its determination of the vertical direction. The practicality of the algorithm for estimating the horizontal decorrelation length of precision errors is shown by application to a set of DEMs produced from images of a desert terrain. Andres Corrada-Emmanuel Machine Learning Photogrammetry Computer Vision http://works.bepress.com/corrada_andres/3 http://works.bepress.com/corrada_andres/3 Sat, 13 Oct 2007 09:02:23 PDT Previous work in social network analysis (SNA) has modeled the existence of links from one entity to another, but not the attributes such as language content or topics on those links. We present the Author-Recipient-Topic (ART) model for social network analysis, which learns topic distributions based on the direction-sensitive messages sent between entities. The model builds on Latent Dirichlet Allocation (LDA) and the Author-Topic (AT) model, adding the key attribute that distribution over topics is conditioned distinctly on both the sender and recipient---steering the discovery of topics according to the relationships between people. We give results on both the Enron email corpus and a researcher's email archive, providing evidence not only that clearly relevant topics are discovered, but that the ART model better predicts people's roles and gives lower perplexity on previously unseen messages. We also present the Role-Author-Recipient-Topic (RART) model, an extension to ART that explicitly represents people's roles. Andrew McCallum Machine Learning Latent Dirichlet Allocation Models http://works.bepress.com/corrada_andres/2 http://works.bepress.com/corrada_andres/2 Fri, 12 Oct 2007 15:55:49 PDT We consider the problem of estimating the vertical uncertainty in Digital Elevation Models when results from asymmetric computer matches are used. The use of asymmetric matches doubles the height estimates available for creating a fused DEM. But if the asymmetric matches are perfectly correlated the variance would not drop by a factor of $1/\sqrt{2}$ as they would for uncorrelated measurements. We present an error model that uses the observed height estimates to measure the average correlation between the asymmetric matches absent any knowledge of the true heights in the DEM. It requires at least three photographs to autonomously estimate the correlation between asymmetric pairs. Experimental results with a specific set of aerial photographs show that the correlation coefficient varies from $0.5$ to $0.9$. This demonstrates that for any algorithm used to fuse DEMs from multiple photographs a better result would be obtained by employing the extra information in asymmetric pairs. Andres Corrada-Emmanuel Machine Learning Photogrammetry Computer Vision http://works.bepress.com/corrada_andres/1 http://works.bepress.com/corrada_andres/1 Fri, 12 Oct 2007 15:55:48 PDT Multiple choice questions (MCQs) are a common data gathering tool. We extend the Latent Dirichlet Allocation (LDA) framework to a collection of MCQ surveys. Topic discovery is turned into group discovery based on survey response patterns. Question choices are equivalent to vocabulary words and are conditioned on the question and the latent group that is used to cluster the survey responders. The structured format of MCQ surveys creates correlations between document `authors' not found in unstructured natural language documents. We demonstrate the utility of the model by considering two performance measures : How well can we predict held-out question answers? What is the discriminatory power of the survey questions? The model should be of interest to anybody that uses MCQ surveys or exams to identify social groups. Andres Corrada-Emmanuel Machine Learning Latent Dirichlet Allocation Models