Comments on: Interesting papers at UAICMOLT 2009 http://hunch.net/?p=813 Machine learning and learning theory research Sun, 07 Feb 2010 15:39:09 -0600 http://wordpress.org/?v=2.8.4 hourly 1 By: Itman http://hunch.net/?p=813&cpage=1#comment-288787 Itman Fri, 24 Jul 2009 08:55:42 +0000 http://hunch.net/?p=813#comment-288787 <i>This paper shows how to directly applying regression in high dimensional vector spaces and have it succeed anyways because the data is naturally low-dimensional.</i> So it is basically a dimensionality-reduction method? Is it a new one? This paper shows how to directly applying regression in high dimensional vector spaces and have it succeed anyways because the data is naturally low-dimensional.

So it is basically a dimensionality-reduction method? Is it a new one?

]]> By: Yisong Yue http://hunch.net/?p=813&cpage=1#comment-286403 Yisong Yue Sun, 12 Jul 2009 18:30:48 +0000 http://hunch.net/?p=813#comment-286403 I also found this paper very interesting: <a href="http://pages.cs.wisc.edu/~jdavis/davisICML09.pdf" rel="nofollow">Deep Transfer via Second-Order Markov Logic</a> by <a href="http://www.cs.washington.edu/homes/jdavis/" rel="nofollow">Jesse Davis</a> and <a href="http://www.cs.washington.edu/homes/pedrod/" rel="nofollow">Pedro Domingos</a>. The authors take a second-order statistical relational learning approach in order to determine the first-order relational regularities across different domains. I also found this paper very interesting:

Deep Transfer via Second-Order Markov Logic by Jesse Davis and Pedro Domingos. The authors take a second-order statistical relational learning approach in order to determine the first-order relational regularities across different domains.

]]>