<?xml version="1.0" encoding="UTF-8"?>

<modsCollection xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.loc.gov/mods/v3" xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-3.xsd">
<mods version="3.3">

<genre>thesis</genre>

<titleInfo><title>Approximating marginals using discrete energy minimization</title></titleInfo>

  
  
<titleInfo type="alternative">
  
  <title>IST Austria Technical Report</title>
</titleInfo>

<note type="publicationStatus">published</note>



<name type="personal">
  <namePart type="given">Filip</namePart>
  <namePart type="family">Korc</namePart>
  <role><roleTerm type="text">author</roleTerm> </role><identifier type="local">476A2FD6-F248-11E8-B48F-1D18A9856A87</identifier></name>
<name type="personal">
  <namePart type="given">Vladimir</namePart>
  <namePart type="family">Kolmogorov</namePart>
  <role><roleTerm type="text">author</roleTerm> </role><identifier type="local">3D50B0BA-F248-11E8-B48F-1D18A9856A87</identifier></name>
<name type="personal">
  <namePart type="given">Christoph</namePart>
  <namePart type="family">Lampert</namePart>
  <role><roleTerm type="text">author</roleTerm> </role><identifier type="local">40C20FD2-F248-11E8-B48F-1D18A9856A87</identifier><description xsi:type="identifierDefinition" type="orcid">0000-0001-8622-7887</description></name>







<name type="corporate">
  <namePart></namePart>
  <identifier type="local">VlKo</identifier>
  <role>
    <roleTerm type="text">department</roleTerm>
  </role>
</name>

<name type="corporate">
  <namePart></namePart>
  <identifier type="local">ChLa</identifier>
  <role>
    <roleTerm type="text">department</roleTerm>
  </role>
</name>








<abstract lang="eng">We consider the problem of inference in agraphical model with binary variables. While in theory it is arguably preferable to compute marginal probabilities, in practice researchers often use MAP inference due to the availability of efficient discrete optimization algorithms. We bridge the gap between the two approaches by introducing the Discrete  Marginals technique in which approximate marginals are obtained by minimizing an objective function with unary and pair-wise terms over a discretized domain. This allows the use of techniques originally devel-oped for MAP-MRF inference and learning. We explore two ways to set up the objective function - by discretizing the Bethe free energy and by learning it  from training data. Experimental results show that for certain types of graphs a learned function can out-perform the  Bethe approximation. We also establish a link between the Bethe free energy and submodular functions.</abstract>

<relatedItem type="constituent">
  <location>
    <url displayLabel="IST-2012-0003_IST-2012-0003.pdf">https://research-explorer.ista.ac.at/download/5396/5490/IST-2012-0003_IST-2012-0003.pdf</url>
  </location>
  <physicalDescription><internetMediaType>application/pdf</internetMediaType></physicalDescription><accessCondition type="restrictionOnAccess">no</accessCondition>
</relatedItem>
<originInfo><publisher>IST Austria</publisher><dateIssued encoding="w3cdtf">2012</dateIssued>
</originInfo>
<language><languageTerm authority="iso639-2b" type="code">eng</languageTerm>
</language>



<relatedItem type="host">
  <identifier type="issn">2664-1690</identifier><identifier type="doi">10.15479/AT:IST-2012-0003</identifier>
<part><extent unit="pages">13</extent>
</part>
</relatedItem>
<relatedItem type="Supplementary material">
  <location>     <url>https://research-explorer.ista.ac.at/record/3124</url>  </location>
</relatedItem>

<extension>
<bibliographicCitation>
<short>F. Korc, V. Kolmogorov, C. Lampert, Approximating Marginals Using Discrete Energy Minimization, IST Austria, 2012.</short>
<mla>Korc, Filip, et al. &lt;i&gt;Approximating Marginals Using Discrete Energy Minimization&lt;/i&gt;. IST Austria, 2012, doi:&lt;a href=&quot;https://doi.org/10.15479/AT:IST-2012-0003&quot;&gt;10.15479/AT:IST-2012-0003&lt;/a&gt;.</mla>
<apa>Korc, F., Kolmogorov, V., &amp;#38; Lampert, C. (2012). &lt;i&gt;Approximating marginals using discrete energy minimization&lt;/i&gt;. IST Austria. &lt;a href=&quot;https://doi.org/10.15479/AT:IST-2012-0003&quot;&gt;https://doi.org/10.15479/AT:IST-2012-0003&lt;/a&gt;</apa>
<ieee>F. Korc, V. Kolmogorov, and C. Lampert, &lt;i&gt;Approximating marginals using discrete energy minimization&lt;/i&gt;. IST Austria, 2012.</ieee>
<chicago>Korc, Filip, Vladimir Kolmogorov, and Christoph Lampert. &lt;i&gt;Approximating Marginals Using Discrete Energy Minimization&lt;/i&gt;. IST Austria, 2012. &lt;a href=&quot;https://doi.org/10.15479/AT:IST-2012-0003&quot;&gt;https://doi.org/10.15479/AT:IST-2012-0003&lt;/a&gt;.</chicago>
<ama>Korc F, Kolmogorov V, Lampert C. &lt;i&gt;Approximating Marginals Using Discrete Energy Minimization&lt;/i&gt;. IST Austria; 2012. doi:&lt;a href=&quot;https://doi.org/10.15479/AT:IST-2012-0003&quot;&gt;10.15479/AT:IST-2012-0003&lt;/a&gt;</ama>
<ista>Korc F, Kolmogorov V, Lampert C. 2012. Approximating marginals using discrete energy minimization, IST Austria, 13p.</ista>
</bibliographicCitation>
</extension>
<recordInfo><recordIdentifier>5396</recordIdentifier><recordCreationDate encoding="w3cdtf">2018-12-12T11:39:06Z</recordCreationDate><recordChangeDate encoding="w3cdtf">2024-10-09T20:54:48Z</recordChangeDate>
</recordInfo>
</mods>
</modsCollection>
