<div dir="ltr">
<div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><div><span style="font-size:12.8px"><font face="verdana, sans-serif"><b>Thesis Defense: Heejin Choi, TTIC</b></font></span><br></div><div style="font-family:arial,helvetica,sans-serif"><span style="font-size:12.8px"><b><br></b></span></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"><div><img src="cid:ii_jle7e2m72_165825d71547a3a3" width="122" height="171" class="gmail-m_5279578848262777243gmail-CToWUd gmail-CToWUd" style="margin-right: 0px;"><br></div><br></div><b style="font-family:arial,helvetica,sans-serif;font-size:12.8px">When:</b><span style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"> Fri</span><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="font-family:arial,helvetica,sans-serif;font-size:12.8px;border-bottom:1px dashed rgb(204,204,204)"><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">day, August 31st at <b>11:00am</b></span></span></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Where: <span> </span></b><span style="font-weight:bold"> </span><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">TTIC, 6045 S. Kenwood Avenue, 5th Floor, Room 526</span></span></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Who: <span> </span></b>Heejin Choi, TTIC </font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><b>Title: </b><span style="font-size:12.8px">Efficient Structured Surrogate Loss and Regularization in Structured Prediction</span></div><br class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline"><div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><div><br></div><div><b><font face="arial, helvetica, sans-serif">Abstract:</font></b></div><div style="text-decoration-style:initial;text-decoration-color:initial">In this dissertation, we focus on several important problems in structured prediction. In <span style="font-size:12.8px">structured prediction, the label has a rich intrinsic substructure, and the loss varies with </span><span style="font-size:12.8px">respect to predicted label and the true label pair. Structured SVM is an extension of binary </span><span style="font-size:12.8px">SVM to adapt to such structured tasks.</span></div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the first part of the dissertation, we study the surrogate losses and its efficient meth<span style="font-size:12.8px">ods. To minimize the empirical risk, a surrogate loss which upper bounds the loss, is used as a proxy to minimize the actual loss. Since the objective function is written in terms of the surrogate loss, the choice of the surrogate loss is important since the performance depends on it. Another issue regarding the surrogate loss is the efficiency of the argmax label inference for the surrogate loss. The efficiency is necessary for the optimization since it is often the most time-consuming step. We present a new class of surrogate losses named a bi-criteria surrogate loss, which is a generalization of the popular surrogate losses. We first investigate an efficient method for a slack rescaling formulation as a starting point utilizing decomposability of the model. Then, we extend the algorithm for the bi-criteria surrogate loss, which is very efficient and shows performance improvement.</span></div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the second part of the dissertation, another important issue of regularization is studied. Specifically, we investigate a problem of regularization in hierarchical classification when a structural imbalance exists in the label structure. We present a method to normalize the structure, as well as a new norm, namely shared Frobenius norm. It is suitable for hierarchical classification that adapts to the data in addition to the label structure.</div><br class="gmail-m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline"></div><br></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><br></div><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><b>Thesis advisor:</b><span> </span><a href="mailto:nati@ttic.edu" target="_blank" style="color:rgb(17,85,204)">Nathan Srebro</a></span><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><br></span></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"> </span></div><br class="gmail-Apple-interchange-newline">
<br><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><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 517</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></div></div></div>
<br><div class="gmail_quote">On Thu, Aug 30, 2018 at 1:54 PM, Mary Marre <span dir="ltr"><<a href="mailto:mmarre@ttic.edu" target="_blank">mmarre@ttic.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">
<div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><div><span style="font-size:12.8px"><font face="verdana, sans-serif"><b>Thesis Defense: Heejin Choi, TTIC</b></font></span><br></div><div style="font-family:arial,helvetica,sans-serif"><span style="font-size:12.8px"><b><br></b></span></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"><div><img src="cid:ii_jle7e2m72_165825d71547a3a3" width="122" height="171" class="m_5279578848262777243gmail-CToWUd" style="margin-right:0px"><br></div><br></div><b style="font-family:arial,helvetica,sans-serif;font-size:12.8px">When:</b><span style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"> Fri</span><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="font-family:arial,helvetica,sans-serif;font-size:12.8px;border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">day, August 31st at <b>11:00am</b></span></span></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Where: <span> </span></b><span style="font-weight:bold"> </span><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">TTIC, 6045 S. Kenwood Avenue, 5th Floor, Room 526</span></span></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Who: <span> </span></b>Heejin Choi, TTIC </font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><b>Title: </b><span style="font-size:12.8px">Efficient Structured Surrogate Loss and Regularization in Structured Prediction</span></div><br class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline"><div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><div><br></div><div><b><font face="arial, helvetica, sans-serif">Abstract:</font></b></div><div style="text-decoration-style:initial;text-decoration-color:initial">In this dissertation, we focus on several important problems in structured prediction. In <span style="font-size:12.8px">structured prediction, the label has a rich intrinsic substructure, and the loss varies with </span><span style="font-size:12.8px">respect to predicted label and the true label pair. Structured SVM is an extension of binary </span><span style="font-size:12.8px">SVM to adapt to such structured tasks.</span></div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the first part of the dissertation, we study the surrogate losses and its efficient meth<span style="font-size:12.8px">ods. To minimize the empirical risk, a surrogate loss which upper bounds the loss, is used as a proxy to minimize the actual loss. Since the objective function is written in terms of the surrogate loss, the choice of the surrogate loss is important since the performance depends on it. Another issue regarding the surrogate loss is the efficiency of the argmax label inference for the surrogate loss. The efficiency is necessary for the optimization since it is often the most time-consuming step. We present a new class of surrogate losses named a bi-criteria surrogate loss, which is a generalization of the popular surrogate losses. We first investigate an efficient method for a slack rescaling formulation as a starting point utilizing decomposability of the model. Then, we extend the algorithm for the bi-criteria surrogate loss, which is very efficient and shows performance improvement.</span></div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the second part of the dissertation, another important issue of regularization is studied. Specifically, we investigate a problem of regularization in hierarchical classification when a structural imbalance exists in the label structure. We present a method to normalize the structure, as well as a new norm, namely shared Frobenius norm. It is suitable for hierarchical classification that adapts to the data in addition to the label structure.</div><br class="m_5279578848262777243gmail-m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline"></div><br></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><br></div><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><b>Thesis advisor:</b><span> </span><a href="mailto:nati@ttic.edu" style="color:rgb(17,85,204)" target="_blank">Nathan Srebro</a></span><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><br></span></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"> </span></div>
<br><div class="gmail_extra"><br clear="all"><div><div class="m_5279578848262777243gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><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 523</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></div>
<br><div class="gmail_quote">On Tue, Aug 28, 2018 at 4:17 PM, Mary Marre <span dir="ltr"><<a href="mailto:mmarre@ttic.edu" target="_blank">mmarre@ttic.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">
<div style="text-decoration-style:initial;text-decoration-color:initial"><div><span style="font-size:12.8px"><font face="verdana, sans-serif"><b>Thesis Defense: Heejin Choi, TTIC</b></font></span><br></div><div style="font-family:arial,helvetica,sans-serif"><span style="font-size:12.8px"><b><br></b></span></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"><div><img src="cid:ii_jle7e2m72_165825d71547a3a3" width="122" height="171" style="margin-right:0px"><br></div><br></div><b style="font-family:arial,helvetica,sans-serif;font-size:12.8px">When:</b><span style="font-family:arial,helvetica,sans-serif;font-size:12.8px;font-weight:bold"> Fri</span><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="font-family:arial,helvetica,sans-serif;font-size:12.8px;border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">day, August 31st at <b>11:00am</b></span></span></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Where: <span> </span></b><span style="font-weight:bold"> </span><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ">TTIC, 6045 S. Kenwood Avenue, 5th Floor, Room 526</span></span></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><b><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aBn" style="border-bottom:1px dashed rgb(204,204,204)"><span class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-m_-8102197176306466350gmail-m_-7224781006463542997gmail-m_5098382079273635224gmail-aQJ"><font face="arial, helvetica, sans-serif"><br></font></span></span></b></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><b>Who: <span> </span></b>Heejin Choi, TTIC </font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><font face="arial, helvetica, sans-serif"><br></font></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><b>Title: </b><span style="font-size:12.8px">Efficient Structured Surrogate Loss and Regularization in Structured Prediction</span></div><br class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline">
<div style="background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial"><div><br></div><div><b><font face="arial, helvetica, sans-serif">Abstract:</font></b></div>
<div style="text-decoration-style:initial;text-decoration-color:initial">In this dissertation, we focus on several important problems in structured prediction. In</div><div style="text-decoration-style:initial;text-decoration-color:initial">structured prediction, the label has a rich intrinsic substructure, and the loss varies with</div><div style="text-decoration-style:initial;text-decoration-color:initial">respect to predicted label and the true label pair. Structured SVM is an extension of binary</div><div style="text-decoration-style:initial;text-decoration-color:initial">SVM to adapt to such structured tasks.</div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the first part of the dissertation, we study the surrogate losses and its efficient meth-</div><div style="text-decoration-style:initial;text-decoration-color:initial">ods. To minimize the empirical risk, a surrogate loss which upper bounds the loss, is used as a proxy to minimize the actual loss. Since the objective function is written in terms of the surrogate loss, the choice of the surrogate loss is important since the performance depends on it. Another issue regarding the surrogate loss is the efficiency of the argmax label inference for the surrogate loss. The efficiency is necessary for the optimization since it is often the most time-consuming step. We present a new class of surrogate losses named a bi-criteria surrogate loss, which is a generalization of the popular surrogate losses. We first investigate an efficient method for a slack rescaling formulation as a starting point utilizing decomposability of the model. Then, we extend the algorithm for the bi-criteria surrogate loss, which is very efficient and shows performance improvement.</div><div style="text-decoration-style:initial;text-decoration-color:initial"><br></div><div style="text-decoration-style:initial;text-decoration-color:initial">In the second part of the dissertation, another important issue of regularization is studied. Specifically, we investigate a problem of regularization in hierarchical classification when a structural imbalance exists in the label structure. We present a method to normalize the structure, as well as a new norm, namely shared Frobenius norm. It is suitable for hierarchical classification that adapts to the data in addition to the label structure.</div><br class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline">
</div><br></div><div style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"><br></div><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><b>Thesis advisor:</b><span> </span><a href="mailto:nati@ttic.edu" target="_blank">Nathan Srebro</a></span><div><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><br></span></div><div><span style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"> </span><br class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail-Apple-interchange-newline">
<br clear="all"><div><div class="m_5279578848262777243m_-7693139298067466964m_-5276955929831589859gmail_signature"><div dir="ltr"><div><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 523</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></div>
</div></div>
</blockquote></div><br></div></div>
</blockquote></div><br></div></div>