<div dir="ltr"><p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal">When: Friday,
December 2nd at noon <span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal">Where: TTIC, 6045 S
Kenwood Avenue, 5th Floor, Room 526<span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal">Who: Srinadh
Bhojanapalli; TTIC<span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">Title:
Dropping Convexity for Faster Low Rank Recovery<span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">Abstract:
Low rank recovery is an important problem in machine learning with applications
in dimensionality reduction, recommendation systems, Euclidean
distance localization, multi-task learning, quantum state tomography
and drug design. Convex optimization methods such as nuclear norm
minimization have been successful in recovering low rank solutions and there
has been lot of research studying its statistical properties. However in
practice, local search methods such as alternating minimization and stochastic
gradient descent on the factored space have been the popular choice for solving
such problems due to their low computational complexity. <span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span> </span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">In
the past couple of years there has been research studying such non-convex
methods and trying to bridge this gap between theory and practice. In this
talk, I will first describe our work studying the effectiveness of gradient
descent for recovering low rank solutions and describe its convergence
properties. Later, I will talk about the global geometric properties of certain
low rank recovery problems that help in recovering the global optima of these
non-convex problems.<span></span></p>
<p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span> </span></p><p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span><br></span></p><div style="font-size:12.8px">******************************<wbr>******************************<wbr>******************************<wbr>**********************</div><div style="font-size:12.8px"><br></div><div style="font-size:12.8px"><div class="gmail_default" style="font-size:12.8px"><b><font face="arial, helvetica, sans-serif">Research at TTIC Seminar Series</font></b></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif"> </font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif">TTIC is hosting a weekly seminar series presenting the research currently underway at the Institute. Every week a different TTIC faculty member will present their research. The lectures are intended both for students seeking research topics and adviser, and for the general TTIC and University of Chicago communities interested in hearing what their colleagues are up to. </font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif"> </font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif">To receive announcements about the seminar series, please subscribe to the mailing list: <a href="https://groups.google.com/a/ttic.edu/group/talks/subscribe" target="_blank">https://groups.google.co<wbr>m/a/ttic.edu/group/<span class="gmail-m_-1056553444316568634gmail-m_593248408351034154gmail-m_8873931552348078312gmail-m_-3900280345830708483gmail-m_2412953322955092339gmail-m_-3568727914368024249gmail-m_-7538648194147727859gmail-il"><span class="gmail-m_-1056553444316568634gmail-m_593248408351034154gmail-m_8873931552348078312gmail-m_-3900280345830708483gmail-m_2412953322955092339gmail-m_-3568727914368024249gmail-il">talks</span></span>/subsc<wbr>ribe</a></font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif"><br></font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif">Speaker details can be found at: <a href="http://www.ttic.edu/tticseminar.php" target="_blank">http://www.ttic.edu/tticse<wbr>minar.php</a>.</font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif"> </font></div><div class="gmail_default" style="font-size:12.8px"><font face="arial, helvetica, sans-serif">For additional questions, please contact Nathan Srebro at <a href="mailto:mcallester@ttic.edu" target="_blank">nati@ttic.edu</a> </font></div><div style="font-size:12.8px"><font face="arial, helvetica, sans-serif"><br></font></div></div><p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span><br></span></p><p class="MsoNormal" style="margin-bottom:0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><span><br></span></p><div><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><font face="arial, helvetica, sans-serif">Mary C. Marre</font><div><font face="arial, helvetica, sans-serif">Administrative Assistant</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">6045 S. Kenwood Avenue</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Room 504</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL 60637</font></i></div><div><i><font face="arial, helvetica, sans-serif">p:(773) 834-1757</font></i></div><div><i><font face="arial, helvetica, sans-serif">f: (773) 357-6970</font></i></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div></div></div></div></div></div>
</div>