<?xml version="1.0" encoding="ISO-8859-1"?><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id>1405-5546</journal-id>
<journal-title><![CDATA[Computación y Sistemas]]></journal-title>
<abbrev-journal-title><![CDATA[Comp. y Sist.]]></abbrev-journal-title>
<issn>1405-5546</issn>
<publisher>
<publisher-name><![CDATA[Instituto Politécnico Nacional, Centro de Investigación en Computación]]></publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id>S1405-55462018000401317</article-id>
<article-id pub-id-type="doi">10.13053/cys-22-4-3078</article-id>
<title-group>
<article-title xml:lang="en"><![CDATA[A Flexible Stochastic Method for Solving the MAP Problem in Topic Models]]></article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Vu]]></surname>
<given-names><![CDATA[Tu]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Bui]]></surname>
<given-names><![CDATA[Xuan]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
<xref ref-type="aff" rid="Aaf"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Than]]></surname>
<given-names><![CDATA[Khoat]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Ichise]]></surname>
<given-names><![CDATA[Ryutaro]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
</contrib-group>
<aff id="Af1">
<institution><![CDATA[,Hanoi University of Science and Technology  ]]></institution>
<addr-line><![CDATA[Hanoi ]]></addr-line>
<country>Vietnam</country>
</aff>
<aff id="Af2">
<institution><![CDATA[,Thai Nguyen University of Information and Communication Technology  ]]></institution>
<addr-line><![CDATA[ ]]></addr-line>
<country>Vietnam</country>
</aff>
<aff id="Af3">
<institution><![CDATA[,National Institute of Informatics  ]]></institution>
<addr-line><![CDATA[Tokyo ]]></addr-line>
<country>Japan</country>
</aff>
<pub-date pub-type="pub">
<day>00</day>
<month>12</month>
<year>2018</year>
</pub-date>
<pub-date pub-type="epub">
<day>00</day>
<month>12</month>
<year>2018</year>
</pub-date>
<volume>22</volume>
<numero>4</numero>
<fpage>1317</fpage>
<lpage>1327</lpage>
<copyright-statement/>
<copyright-year/>
<self-uri xlink:href="http://www.scielo.org.mx/scielo.php?script=sci_arttext&amp;pid=S1405-55462018000401317&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.org.mx/scielo.php?script=sci_abstract&amp;pid=S1405-55462018000401317&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.org.mx/scielo.php?script=sci_pdf&amp;pid=S1405-55462018000401317&amp;lng=en&amp;nrm=iso"></self-uri><abstract abstract-type="short" xml:lang="en"><p><![CDATA[Abstract: The estimation of the posterior distribution is the core problem in topic models, unfortunately it is intractable. There are approximation and sampling methods proposed to solve it. However, most of them do not have any clear theoretical guarantee of neither quality nor rate of convergence. Online Maximum a Posteriori Estimation (OPE) is another approach with concise guarantee on quality and convergence rate, in which we cast the estimation of the posterior distribution into a non-convex optimization problem. In this paper, we propose a more general and flexible version of OPE, namely Generalized Online Maximum a Posteriori Estimation (G-OPE), which not only enhances the flexibility of OPE in different real-world situations but also preserves key advantage theoretical characteristics of OPE when comparing to the state-of-the-art methods. We employ G-OPE as inference a document within large text corpora. The experimental and theoretical results show that our new approach performs better than OPE and other state-of-the-art methods.]]></p></abstract>
<kwd-group>
<kwd lng="en"><![CDATA[Topic models]]></kwd>
<kwd lng="en"><![CDATA[posterior inference]]></kwd>
<kwd lng="en"><![CDATA[online MAP estimation]]></kwd>
<kwd lng="en"><![CDATA[large-scale learning]]></kwd>
<kwd lng="en"><![CDATA[non-convex optimization]]></kwd>
</kwd-group>
</article-meta>
</front><back>
<ref-list>
<ref id="B1">
<label>1</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Aletras]]></surname>
<given-names><![CDATA[N.]]></given-names>
</name>
<name>
<surname><![CDATA[Stevenson]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<source><![CDATA[Evaluating topic coherence using distributional semantics]]></source>
<year>2013</year>
<conf-name><![CDATA[ 10th International Conference on Computational Semantics]]></conf-name>
<conf-loc> </conf-loc>
<page-range>13-22</page-range></nlm-citation>
</ref>
<ref id="B2">
<label>2</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Asuncion]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
<name>
<surname><![CDATA[Welling]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
<name>
<surname><![CDATA[Smyth]]></surname>
<given-names><![CDATA[P.]]></given-names>
</name>
<name>
<surname><![CDATA[Teh]]></surname>
<given-names><![CDATA[Y. W.]]></given-names>
</name>
</person-group>
<source><![CDATA[On smoothing and inference for topic models]]></source>
<year>2009</year>
<conf-name><![CDATA[ 25th Conference on Uncertainty in Artificial Intelligence]]></conf-name>
<conf-loc> </conf-loc>
<page-range>27-34</page-range></nlm-citation>
</ref>
<ref id="B3">
<label>3</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Blei]]></surname>
<given-names><![CDATA[D. M.]]></given-names>
</name>
<name>
<surname><![CDATA[Ng]]></surname>
<given-names><![CDATA[A. Y.]]></given-names>
</name>
<name>
<surname><![CDATA[Jordan]]></surname>
<given-names><![CDATA[M. I.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Latent dirichlet allocation]]></article-title>
<source><![CDATA[Journal of machine Learning research]]></source>
<year>2003</year>
<volume>3</volume>
<numero>Jan</numero>
<issue>Jan</issue>
<page-range>993-1022</page-range></nlm-citation>
</ref>
<ref id="B4">
<label>4</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Bouma]]></surname>
<given-names><![CDATA[G]]></given-names>
</name>
</person-group>
<source><![CDATA[Normalized (pointwise) mutual information in collocation extraction]]></source>
<year>2009</year>
<page-range>31-40</page-range><publisher-name><![CDATA[German Society for Computational Linguistics &amp; Language Technology]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B5">
<label>5</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Dempster]]></surname>
<given-names><![CDATA[A. P.]]></given-names>
</name>
<name>
<surname><![CDATA[Laird]]></surname>
<given-names><![CDATA[N. M.]]></given-names>
</name>
<name>
<surname><![CDATA[Rubin]]></surname>
<given-names><![CDATA[D. B.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Maximum likelihood from incomplete data via the em algorithm]]></article-title>
<source><![CDATA[Journal of the royal statistical society. Series B]]></source>
<year>1977</year>
<page-range>1-38</page-range></nlm-citation>
</ref>
<ref id="B6">
<label>6</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Feller]]></surname>
<given-names><![CDATA[W]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[The general form of the so-called law of the iterated logarithm]]></article-title>
<source><![CDATA[Transactions of the American Mathematical Society]]></source>
<year>1943</year>
<volume>54</volume>
<numero>3</numero>
<issue>3</issue>
<page-range>373-402</page-range></nlm-citation>
</ref>
<ref id="B7">
<label>7</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Frank]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
<name>
<surname><![CDATA[Wolfe]]></surname>
<given-names><![CDATA[P.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[An algorithm for quadratic programming]]></article-title>
<source><![CDATA[Naval Research Logistics]]></source>
<year>1956</year>
<volume>3</volume>
<numero>1-2</numero>
<issue>1-2</issue>
<page-range>95-110</page-range></nlm-citation>
</ref>
<ref id="B8">
<label>8</label><nlm-citation citation-type="book">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Griffiths]]></surname>
<given-names><![CDATA[T. L.]]></given-names>
</name>
<name>
<surname><![CDATA[Steyvers]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Finding scientific topics]]></article-title>
<source><![CDATA[Proceedings of the National Academy of Sciences]]></source>
<year>2004</year>
<volume>101</volume>
<page-range>5228-35</page-range><publisher-name><![CDATA[National Acad Sciences]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B9">
<label>9</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Hazan]]></surname>
<given-names><![CDATA[E.]]></given-names>
</name>
<name>
<surname><![CDATA[Kale]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
</person-group>
<source><![CDATA[Projection-free online learning]]></source>
<year>2012</year>
<conf-name><![CDATA[ 29th International Con-ference on International Conference on Machine Learning]]></conf-name>
<conf-loc> </conf-loc>
<page-range>1843-50</page-range></nlm-citation>
</ref>
<ref id="B10">
<label>10</label><nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Hoffman]]></surname>
<given-names><![CDATA[M. D.]]></given-names>
</name>
<name>
<surname><![CDATA[Blei]]></surname>
<given-names><![CDATA[D. M.]]></given-names>
</name>
<name>
<surname><![CDATA[Wang]]></surname>
<given-names><![CDATA[C.]]></given-names>
</name>
<name>
<surname><![CDATA[Paisley]]></surname>
<given-names><![CDATA[J. W.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Stochastic variational inference]]></article-title>
<source><![CDATA[Journal of machine Learning research]]></source>
<year>2013</year>
<volume>14</volume>
<numero>1</numero>
<issue>1</issue>
<page-range>1303-47</page-range></nlm-citation>
</ref>
<ref id="B11">
<label>11</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Lau]]></surname>
<given-names><![CDATA[J. H.]]></given-names>
</name>
<name>
<surname><![CDATA[Newman]]></surname>
<given-names><![CDATA[D.]]></given-names>
</name>
<name>
<surname><![CDATA[Baldwin]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
</person-group>
<source><![CDATA[Machine reading tea leaves: Automatically evaluating topic coherence and topic model quality]]></source>
<year>2014</year>
<conf-name><![CDATA[ 14th Conference of the European Chapter of the Association for Computational Linguistics]]></conf-name>
<conf-loc> </conf-loc>
<page-range>530-9</page-range></nlm-citation>
</ref>
<ref id="B12">
<label>12</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Mimno]]></surname>
<given-names><![CDATA[D.]]></given-names>
</name>
<name>
<surname><![CDATA[Hoffman]]></surname>
<given-names><![CDATA[M. D.]]></given-names>
</name>
<name>
<surname><![CDATA[Blei]]></surname>
<given-names><![CDATA[D. M.]]></given-names>
</name>
</person-group>
<source><![CDATA[Sparse stochastic inference for latent dirichlet allocation]]></source>
<year>2012</year>
<conf-name><![CDATA[ 29th International Coference on International Conference on Machine Learning]]></conf-name>
<conf-loc> </conf-loc>
<page-range>1515-22</page-range></nlm-citation>
</ref>
<ref id="B13">
<label>13</label><nlm-citation citation-type="confpro">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Reddi]]></surname>
<given-names><![CDATA[S. J.]]></given-names>
</name>
<name>
<surname><![CDATA[Sra]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
<name>
<surname><![CDATA[Póczos]]></surname>
<given-names><![CDATA[B.]]></given-names>
</name>
<name>
<surname><![CDATA[J.Smola]]></surname>
<given-names><![CDATA[A.]]></given-names>
</name>
</person-group>
<source><![CDATA[Stochastic frank-wolfe methods for non-convex optimization]]></source>
<year>2016</year>
<conf-name><![CDATA[ 54th Annual Allerton Conference on Communication, Control, and Computing]]></conf-name>
<conf-loc> </conf-loc>
<page-range>1244-51</page-range></nlm-citation>
</ref>
<ref id="B14">
<label>14</label><nlm-citation citation-type="">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Sontag]]></surname>
<given-names><![CDATA[D.]]></given-names>
</name>
<name>
<surname><![CDATA[Roy]]></surname>
<given-names><![CDATA[D.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Complexity of inference in latent dirichlet allocation]]></article-title>
<source><![CDATA[Advances in neural information processing systems]]></source>
<year>2011</year>
<page-range>1008-16</page-range></nlm-citation>
</ref>
<ref id="B15">
<label>15</label><nlm-citation citation-type="">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Teh]]></surname>
<given-names><![CDATA[Y. W.]]></given-names>
</name>
<name>
<surname><![CDATA[Kurihara]]></surname>
<given-names><![CDATA[K.]]></given-names>
</name>
<name>
<surname><![CDATA[Welling]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Collapsed variational inference for hdp]]></article-title>
<source><![CDATA[Proceedings of Advances in Neural Information Processing Systems]]></source>
<year>2007</year>
<page-range>1481-8</page-range></nlm-citation>
</ref>
<ref id="B16">
<label>16</label><nlm-citation citation-type="">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Than]]></surname>
<given-names><![CDATA[K.]]></given-names>
</name>
<name>
<surname><![CDATA[Doan]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
</person-group>
<source><![CDATA[Guaranteed inference in topic models]]></source>
<year>2015</year>
</nlm-citation>
</ref>
</ref-list>
</back>
</article>
