<!DOCTYPE html>
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body dir="ltr">
    <div class="moz-forward-container">
      <div class="moz-forward-container">
        <div class="moz-forward-container">
          <div lang="EN-US" link="blue" vlink="purple"
            style="word-wrap:break-word">
            <div>
              <div class="x_WordSection1">
                <div>
                  <p class="v1v1msonormal"><span
                      style="font-size:11.0pt">Dear AI scientists and
                      students,<span style="color:black"></span></span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt">University of Connecticut
                      hosts the <span style="color:black">attached
                        upcoming exciting lecture by Dr. Ioannis Pitas <i>Chair
                          of the International AI Doctoral Academy (</i></span><a
                        href="https://www.i-aida.org/"
                        originalsrc="https://www.i-aida.org/"
                        target="_blank" moz-do-not-send="true"><i>AIDA</i></a><i><span
                          style="color:black">) Principal Researcher at
                          Aristotle University of Thessaloniki (AUTH)
                          and CERTH/ITI Greece.</span></i> You are
                      welcomed to attend.</span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt; color:black">When &
                      Where: June 11th, 2026 @ 12:00pm </span><span
                      style="font-size:11.0pt">EDT, <span
                        style="color:black">IPB Room 203</span>,
                      University of Connecticut, Storrs, CT, USA<span
                        style="color:black"></span></span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt; color:black">Virtual Zoom
                      link: </span><span
                      style="font-size:11.0pt; color:#1264A3"><a
                        href="https://authgr.zoom.us/j/92451144378"
originalsrc="https://authgr.zoom.us/j/92451144378"
                        moz-do-not-send="true"
                        class="moz-txt-link-freetext">https://authgr.zoom.us/j/92451144378</a></span><span
                      style="font-size:11.0pt"></span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt; color:#1264A3"> </span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><b><span
                        style="font-size:11.0pt; color:black">Decentralized
                        Machine Learning for Natural Disaster Management
                        Applications</span></b><span
                      style="font-size:11.0pt; color:black"></span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><b><span
                        style="font-size:11.0pt; color:black">Abstract:</span></b><span
                      style="font-size:11.0pt; color:black"></span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt; color:black">This lecture
                      overviews decentralized and distributed DNN
                      architectures and their implementation in
                      cloud/edge environments. Big data analysis can be
                      greatly facilitated if decentralized/distributed
                      DNN architectures are employed that interact with
                      each other for DNN training and/or inference using
                      the human Teacher-Student education paradigm. A
                      novel Learning-by-Education Node Community (LENC)
                      framework is presented that facilitates
                      communication and knowledge exchange among diverse
                      Deep Neural Networks (DNN) agents, undertaking the
                      role of a student or teacher DNN by offering or
                      absorbing knowledge respectively. The framework
                      enables efficient and effective knowledge transfer
                      among participating DNN agents while enhancing
                      their learning capabilities and fostering their
                      collaboration among diverse networks. The proposed
                      framework addresses the challenges of handling
                      diverse training data distributions and the
                      limitations of individual DNN agent learning
                      abilities. The LENC framework ensures the
                      exploitation of the best available teacher
                      knowledge upon learning a new task and protects
                      the DNN agents from catastrophic forgetting. The
                      experiments demonstrate the LENC framework
                      functionalities on multiple teacher-student
                      learning techniques and their integration with
                      lifelong learning. Our experiments manifest the
                      LEMA framework’s ability to maximize the accuracy
                      of all participating DNN agents in classification
                      tasks by leveraging the collaborative knowledge of
                      the framework. A LENC framework implementation in
                      cloud/edge environments is also overviewed.</span></p>
                </div>
                <div style="margin-top:12.0pt; margin-bottom:12.0pt">
                  <p class="x_MsoNormal" style="line-height:15.0pt"><span
                      style="font-size:11.0pt; color:black">Applications
                      are presented in big visual data analysis tasks
                      for Natural Disaster Management (NDM), focusing on
                      flood and wildfire monitoring, combatting and
                      management. As many players enter into NDM, from
                      first responders to civil authorities, NDM needs
                      big data analysis that is typically decentralized
                      and DDN-based and resides in the entire
                      edge-to-cloud computing continuum.</span></p>
                </div>
              </div>
            </div>
          </div>
        </div>
      </div>
    </div>
  </body>
</html>