BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Repeated Motif Hierarchical Stochastic Blockmodels - Carey Priebe 
 (Johns Hopkins University)
DTSTART:20160715T133000Z
DTEND:20160715T140000Z
UID:TALK66777@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span><span>Co-authors: Vince Lyzinski (Johns Hopkins  Univers
 ity)\, Minh Tang (Johns Hopkins University)\, Avanti Athreya  (Johns Hopki
 ns University)\, Youngser Park (Johns Hopkins  University)\, Joshua Vogels
 tein (Johns Hopkins University)\, Keith  Levin (Johns Hopkins University) 
 <br></span> <br>Based on our methodology for community detection and commu
 nity comparison in  graphs (Lyzinski et al.\, 2015\, <a href="http://arxiv
 .org/abs/1503.02115" target="_blank" rel="nofollow">http://arxiv.org/abs/1
 503.02115</a>)\, we formulate a  model selection procedure for deciding wh
 ether a hierarchical stochastic  blockmodel graph supports the conjecture 
 of repeated motifs. Such a graph  inference procedure promises to address 
 a fundamental outstanding question  regarding the atoms of neural computat
 ion (Marcus\, Marblestone & Dean\,  2014\, <a href="http://arxiv.org/abs/1
 410.8826" target="_blank" rel="nofollow">http://arxiv.org/abs/1410.8826</a
 >)</span><span>. <br>Related Links </span><ul> <li><a target="_blank" rel=
 "nofollow">http://www.cis.jhu.edu/~parky/RMHSBM/rmhsbm.html</a></li></ul>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR