PUMA publications for /tag/kallimachos%20learninghttps://puma.uni-kassel.de/tag/kallimachos%20learningPUMA RSS feed for /tag/kallimachos%20learning2024-03-29T03:00:43+01:00Dynamic Auto-Encoders for Semantic Indexinghttps://puma.uni-kassel.de/bibtex/2fc3e0e3af595f9a46df6bc9233df836f/hothohotho2015-04-21T14:10:12+02:00deep kallimachos lda learning model toread <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Piotr Mirowski" itemprop="url" href="/author/Piotr%20Mirowski"><span itemprop="name">P. Mirowski</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Marc'Aurelio Ranzato" itemprop="url" href="/author/Marc'Aurelio%20Ranzato"><span itemprop="name">M. Ranzato</span></a></span>, and <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Yann LeCun" itemprop="url" href="/author/Yann%20LeCun"><span itemprop="name">Y. LeCun</span></a></span>. </span><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"></span>(<em><span>2010<meta content="2010" itemprop="datePublished"/></span></em>)Tue Apr 21 14:10:12 CEST 2015Dynamic Auto-Encoders for Semantic Indexing2010deep kallimachos lda learning model toread Neuer TabLarge-scale Multi-label Learning with Missing Labelshttps://puma.uni-kassel.de/bibtex/2716e5270c1dcb3a1e4eedf9934859021/hothohotho2015-03-03T18:17:32+01:00classification kallimachos label large learning multi <span class="authorEditorList"><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Hsiang-Fu Yu" itemprop="url" href="/author/Hsiang-Fu%20Yu"><span itemprop="name">H. Yu</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Prateek Jain" itemprop="url" href="/author/Prateek%20Jain"><span itemprop="name">P. Jain</span></a></span>, <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Purushottam Kar" itemprop="url" href="/author/Purushottam%20Kar"><span itemprop="name">P. Kar</span></a></span>, and <span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Inderjit S. Dhillon" itemprop="url" href="/author/Inderjit%20S.%20Dhillon"><span itemprop="name">I. Dhillon</span></a></span>. </span>(<em><span>2013<meta content="2013" itemprop="datePublished"/></span></em>)<em>cite arxiv:1307.5101.</em>Tue Mar 03 18:17:32 CET 2015cite arxiv:1307.5101Large-scale Multi-label Learning with Missing Labels2013classification kallimachos label large learning multi The multi-label classification problem has generated significant interest in
recent years. However, existing approaches do not adequately address two key
challenges: (a) the ability to tackle problems with a large number (say
millions) of labels, and (b) the ability to handle data with missing labels. In
this paper, we directly address both these problems by studying the multi-label
problem in a generic empirical risk minimization (ERM) framework. Our
framework, despite being simple, is surprisingly able to encompass several
recent label-compression based methods which can be derived as special cases of
our method. To optimize the ERM problem, we develop techniques that exploit the
structure of specific loss functions - such as the squared loss function - to
offer efficient algorithms. We further show that our learning framework admits
formal excess risk bounds even in the presence of missing labels. Our risk
bounds are tight and demonstrate better generalization performance for low-rank
promoting trace-norm regularization when compared to (rank insensitive)
Frobenius norm regularization. Finally, we present extensive empirical results
on a variety of benchmark datasets and show that our methods perform
significantly better than existing label compression based methods and can
scale up to very large datasets such as the Wikipedia dataset.Large-scale Multi-label Learning with Missing Labels