<!DOCTYPE html>
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>Hi Everyone,</p>
    <p>This is starting right now in GHC 8102.  Please come and join to
      hear about UQ!</p>
    <p>Jeff.</p>
    <p><br>
    </p>
    <div class="moz-forward-container"><br>
      <br>
      -------- Forwarded Message --------
      <table cellpadding="0" cellspacing="0" border="0"
        class="moz-email-headers-table">
        <tbody>
          <tr>
            <th valign="BASELINE" align="RIGHT" nowrap="nowrap">Subject:
            </th>
            <td>Reminder - Thesis Defense - July 23, 2025 - Youngseog
              Chung - Methods for Calibrated Uncertainty Quantification
              and Understanding its Utility</td>
          </tr>
          <tr>
            <th valign="BASELINE" align="RIGHT" nowrap="nowrap">Date: </th>
            <td>Wed, 23 Jul 2025 13:32:49 -0400</td>
          </tr>
          <tr>
            <th valign="BASELINE" align="RIGHT" nowrap="nowrap">From: </th>
            <td>Diane L Stidle <a class="moz-txt-link-rfc2396E" href="mailto:stidle@andrew.cmu.edu"><stidle@andrew.cmu.edu></a></td>
          </tr>
          <tr>
            <th valign="BASELINE" align="RIGHT" nowrap="nowrap">To: </th>
            <td><a class="moz-txt-link-abbreviated" href="mailto:ml-seminar@cs.cmu.edu">ml-seminar@cs.cmu.edu</a> <a class="moz-txt-link-rfc2396E" href="mailto:ML-SEMINAR@cs.cmu.edu"><ML-SEMINAR@cs.cmu.edu></a>,
              <a class="moz-txt-link-abbreviated" href="mailto:jsnoek@google.com">jsnoek@google.com</a></td>
          </tr>
        </tbody>
      </table>
      <br>
      <br>
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div>
          <div dir="ltr">
            <div>
              <div><b><i>Thesis Defense</i></b></div>
              <div><br>
              </div>
              <div>
                <div>Date: July 23, 2025 (Wed)<br>
                </div>
                <div>Time: 2:30pm (ET)<br>
                </div>
                <div>Place: GHC 8102 & Remote<br>
                </div>
                <div>PhD Candidate: Youngseog Chung</div>
              </div>
              <div>
                <div><b><br>
                  </b></div>
                <div><b>Thesis Title:</b><br>
                  Methods for Calibrated Uncertainty Quantification and
                  Understanding its Utility<br>
                </div>
                <div><b><br>
                  </b></div>
                <div><b>Abstract:</b> </div>
                <div>As machine learning models have become more capable
                  of dealing with complex data, they have been entrusted
                  with an increasing array of predictive tasks. With
                  such growing reliance on model predictions, being able
                  to assess whether a given model prediction is reliable
                  has become equally important. Uncertainty
                  quantification (UQ) plays a critical role in this
                  context by providing a measure of confidence in a
                  model's predictions. In this thesis, I address the
                  problem of UQ in machine learning in three different
                  stages.<br>
                  <br>
                </div>
                <div>The first section presents an overview of
                  evaluation in UQ and an open-source software which
                  provides various utilities in evaluating, visualizing,
                  and recalibrating predictive uncertainty. The second
                  section discusses algorithms designed to produce
                  well-calibrated predictive uncertainties in regression
                  models, which output a distribution over
                  continuous-valued outputs. The first work in this
                  section presents a suite of algorithms for training
                  univariate probabilistic regression models, and the
                  second work discusses an extension to the multivariate
                  setting. The third section presents the utilization of
                  predictive uncertainties in the decision-making
                  setting. The application setting dictates how the
                  uncertainties will be used, and I present a collection
                  of works which utilizes uncertainties in the
                  single-step decision making setting, sequential
                  decision-making setting, and in model-based
                  reinforcement learning.</div>
                <div><b><br>
                    Thesis Committee:</b></div>
                <div>Jeff Schneider (Chair) </div>
                <div>Aarti Singh</div>
                <div>Zico Kolter </div>
                <div>Jasper Snoek (Google Deepmind)</div>
                <div><b><br>
                  </b></div>
                <div><b>Link to the draft document: </b></div>
                <div><a
href="https://youngseogchung.github.io/docs/Thesis.pdf" target="_blank"
                    moz-do-not-send="true" class="moz-txt-link-freetext">https://youngseogchung.github.io/docs/Thesis.pdf</a><br>
                </div>
                <div><br>
                </div>
                <div><b>Zoom meeting link:</b> </div>
                <div><a
href="https://cmu.zoom.us/j/99486342780?pwd=HQOzBqHgdHLLDbbQonw0nbgV4ngvQy.1&jst=2"
                    target="_blank" moz-do-not-send="true">https://cmu.zoom.us/j/99486342780?pwd=HQOzBqHgdHLLDbbQonw0nbgV4ngvQy.1&jst=2</a></div>
              </div>
            </div>
          </div>
          <div dir="ltr"><br>
          </div>
          <br clear="all">
        </div>
        <br>
        <span class="gmail_signature_prefix">-- </span><br>
        <div dir="ltr" class="gmail_signature"
          data-smartmail="gmail_signature">
          <div dir="ltr"><span>-- </span><br>
            <div>Diane Stidle <br>
              PhD Program Manager <br>
              Machine Learning Department <br>
              Carnegie Mellon University <br>
              <a href="mailto:diane@cs.cmu.edu" target="_blank"
                moz-do-not-send="true" class="moz-txt-link-freetext">diane@cs.cmu.edu</a></div>
          </div>
        </div>
      </div>
    </div>
  </body>
</html>