<div dir="ltr">A reminder about Yichong's defense coming up this Monday at 9am.<div><br><div>All details below, it will be a fun talk so please come and join if you can!</div><div><br></div><div>Cheers,</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="auto"><<a href="mailto:stidle@andrew.cmu.edu">stidle@andrew.cmu.edu</a>></span><br>Date: Mon, Jun 15, 2020 at 1:49 PM<br>Subject: Thesis Defense - June 29, 2020 - Yichong Xu - Learning and Decision Making from Diverse Forms of Information<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:jcl@microsoft.com">jcl@microsoft.com</a>><br></div><br><br>
<div>
<p><i><b>Thesis Defense</b></i></p>
<p>Date: June 29, 2020<br>
Time: 9:00am (EDT)<br>
PhD Candidate: Yichong Xu</p>
<p>Virtual Presentation Link:<br>
<a href="https://cmu.zoom.us/j/99909151454" target="_blank">https://cmu.zoom.us/j/99909151454</a></p>
<p><font size="-1"><b>Title: </b><span style="font-weight:bold;background-color:white" lang="en-US"></span><span style="font-weight:bold" lang="zh-CN">Learning and Decision Making from Diverse Forms
of Information</span></font></p>
<p><font size="-1">Abstract:<br>
Classical machine learning posits that data are independently
and identically distributed, in a single format usually the same
as test data. In modern applications however, additional
information in other formats might be available freely or at a
lower cost. For example, in data crowdsourcing we can collect
preferences over the data points instead of directly asking the
labels of a single data point at a lower cost. In natural
language understanding problems, we might have limited amount of
data in the target domain, but can use a large amount of general
domain data for free.</font>
</p>
<div style="margin:0in;font-size:9pt">
<p><font size="-1">The main topic of this thesis is to study how
to efficiently incorporate these diverse forms of information
into the learning and decision making process. We study two
representative paradigms in this thesis. Firstly, we study
learning and decision making problems with direct labels and
comparisons. Our algorithms can efficiently combine
comparisons with direct labels so that the total learning cost
can be greatly reduced. Secondly, we study multi-task learning
problems from multiple domain data, and design algorithms to
transfer the data from a general, abundant domain to the
target domain. We show theoretical guarantees of our
algorithms as well as their statistical minimaxity through
information-theoretic limits. On the practical side, we
demonstrate promising experimental results on price estimation
and natural language understanding tasks.</font></p>
</div>
<p>
</p>
<div style="margin:0in;font-size:9pt">
<p><font size="-1"><b>Thesis Committee:</b><br>
Artur Dubrawski, Co-Chair <br>
Aarti Singh, Co-Chair <br>
Sivaraman Balakrishnan<br>
John Langford (Microsoft Research)</font></p>
</div>
<pre cols="72">--
Diane Stidle
Graduate Programs Manager
Machine Learning Department
Carnegie Mellon University
<a href="mailto:stidle@andrew.cmu.edu" target="_blank">stidle@andrew.cmu.edu</a>
412-268-1299</pre>
</div>
</div></div></div></div>