<div dir="ltr">Team,<div><br></div><div>Please come and cheer one of our own when they are making a decisive step towards becoming "Doctor Lei"!</div><div><br></div><div>It will be practically next door (NSH 3002) at 10am this Wednesday.</div><div><br></div><div>Artur<br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">---------- Forwarded message ---------<br>From: <strong class="gmail_sendername" dir="auto">Diane Stidle</strong> <span dir="ltr"><<a href="mailto:stidle@andrew.cmu.edu">stidle@andrew.cmu.edu</a>></span><br>Date: Mon, May 13, 2019 at 10:26 AM<br>Subject: Thesis Proposal - May 15th - Eric Lei - Multi-View Relationships for Analytics and Inference<br>To: <a href="mailto:ml-seminar@cs.cmu.edu">ml-seminar@cs.cmu.edu</a> <<a href="mailto:ML-SEMINAR@cs.cmu.edu">ML-SEMINAR@cs.cmu.edu</a>>, <<a href="mailto:labov1@llnl.gov">labov1@llnl.gov</a>>, Mario Berges <<a href="mailto:marioberges@cmu.edu">marioberges@cmu.edu</a>><br></div><br><br>
<div>
<p><i><b>Thesis Proposal</b></i></p>
<p>Date: May 15, 2019<br>
Time: 10:00am<br>
Place: NSH 3002<br>
Speaker: Eric Lei</p>
<p><b>Title: Multi-View Relationships for Analytics and Inference</b></p>
<p>Abstract:<br>
An interesting area of machine learning is methods for multi-view
data. Classical multi-view literature typically exploits multiple
views to reduce noise in data via conditional independence.
However, the multi-view relationships themselves are underutilized
as factors for analyzing the data. In this work, we investigate
the usefulness of these relationships in descriptive analytics and
inference. In Part I, we cover a problem in radiation detection
about inference of latent variables that are shared dependencies
of multiple views. We show how the views can be aggregated by
filtering their inferences collectively using domain knowledge
about their relationships. In Part II, we address the problem of
modeling multi-view structure directly and applying such structure
to descriptive analytics and inference. We propose a method for
radiation detection that learns linear multi-view correlations and
detects anomalies when these correlations are disrupted. Next, we
extend this idea to nonlinear multi-view correlations by
introducing a clustering method whose correlations are
cluster-wise linear. This method is applied to perform descriptive
analytics on a medical problem. Furthermore, we extend this work
to classification and demonstrate it on a load monitoring problem.
Lastly, we propose an extension to regression. <br>
</p>
<div><b>Thesis Committee:</b> <br>
Artur Dubrawski (Chair)<br>
Barnabas Poczos<br>
Mario Berges<br>
Simon Labov (Lawrence Livermore National Laboratory)</div>
<div><br>
<div dir="ltr">Link to draft thesis document: <a href="https://www.dropbox.com/s/uhl0swsodthsbw3/main1.pdf?dl=0" target="_blank">https://www.dropbox.com/s/uhl0swsodthsbw3/main1.pdf?dl=0</a></div>
</div>
<pre class="m_-4843605697981084267moz-signature" cols="72">--
Diane Stidle
Graduate Programs Manager
Machine Learning Department
Carnegie Mellon University
<a class="m_-4843605697981084267moz-txt-link-abbreviated" href="mailto:stidle@cmu.edu" target="_blank">stidle@cmu.edu</a>
412-268-1299</pre>
</div>
</div></div></div>