<!DOCTYPE html>
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <div class="moz-forward-container">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <br>
      <div class="moz-forward-container">
        <div class="moz-forward-container">
          <div class="moz-forward-container">
            <div id="website_title">
              <div><b>DEEPK 2024<br>
                  <i>International Workshop on Deep Learning and Kernel
                    Machines</i></b><br>
              </div>
              <div><br>
              </div>
              <div>March 7-8, 2024, Leuven, Arenberg Castle, Belgium<br>
                <a class="moz-txt-link-freetext"
href="https://www.esat.kuleuven.be/stadius/E/DEEPK2024"
                  moz-do-not-send="true">https://www.esat.kuleuven.be/stadius/E/DEEPK2024</a></div>
              <div><br>
              </div>
              <div><b><i>- Main scope -</i></b><br>
              </div>
              <div><br>
              </div>
              <div>Major progress and impact has been achieved through
                deep learning architectures with many exciting
                applications such as by generative models and
                transformers. At the same time it triggers new questions
                on the fundamental possibilities and limitations of the
                models, with respect to representations, scalability,
                learning and generalization aspects. Through
                kernel-based methods often a deeper understanding and
                solid foundations have been obtained, complementary to
                the powerful and flexible deep learning architectures.
                Recent examples are understanding generalization of
                over-parameterized models in the double descent
                phenomenon and conceiving attention mechanisms in
                transformers as kernel machines. The aim of DEEPK 2024
                is to provide a multi-disciplinary forum where
                researchers of different communities can meet, to find
                new synergies between deep learning and kernel machines,
                both at the level of theory and applications. <br>
              </div>
              <div><br>
              </div>
              <div><b><i>- Topics - </i></b><br>
              </div>
              <div><br>
              </div>
              <div>Topics include but are not limited to:<br>
                <ul>
                  <li>Deep learning and generalization </li>
                  <li>Double descent phenomenon and over-parameterized
                    models </li>
                  <li>Transformers and asymmetric kernels </li>
                  <li>Attention mechanisms, kernel singular value
                    decomposition </li>
                  <li>Learning with asymmetric kernels </li>
                  <li>Duality and deep learning </li>
                  <li>Regularization schemes, normalization </li>
                  <li>Neural tangent kernel </li>
                  <li>Deep learning and Gaussian processes </li>
                  <li>Transformers, support vector machines and least
                    squares support vector machines </li>
                  <li>Autoencoders, neural networks and kernel methods </li>
                  <li>Kernel methods in GANs, variational autoencoders,
                    diffusion models, Generative Flow Networks </li>
                  <li>Generative kernel machines </li>
                  <li>Deep Kernel PCA, deep kernel machines, deep
                    eigenvalues, deep eigenvectors </li>
                  <li>Restricted Boltzmann machines, Restricted kernel
                    machines, deep learning, energy based models </li>
                  <li>Disentanglement and explainability </li>
                  <li>Tensors, kernels and deep learning </li>
                  <li>Convolutional kernels </li>
                  <li>Sparsity, robustness, low-rank representations,
                    compression </li>
                  <li>Nystrom method, Nystromformer </li>
                  <li>Efficient training methods </li>
                  <li>Lagrange duality, Fenchel duality, estimation in
                    Hilbert spaces, reproducing kernel Hilbert spaces,
                    vector-valued reproducing kernel Hilbert spaces,
                    Krein spaces, Banach spaces, RKHS and C*-algebra</li>
                  <li>Applications</li>
                </ul>
                <br>
              </div>
              <div><b><i>- Invited Speakers -</i></b><br>
              </div>
              <div>
                <ul>
                  <li><a href="http://misha.belkin-wang.org/"
                      moz-do-not-send="true">Mikhail Belkin</a>
                    (University of California San Diego)<br>
                  </li>
                  <li><a href="https://www.epfl.ch/labs/lions/"
                      moz-do-not-send="true">Volkan Cevher</a> (EPFL)<br>
                  </li>
                  <li><a
                      href="https://perso.telecom-paristech.fr/fdalche/"
                      moz-do-not-send="true">Florence d'Alche-Buc</a><a
                      moz-do-not-send="true"> (Telecom Paris, Institut
                      Polytechnique de Paris)<br>
                    </a></li>
                  <li><a href="https://lear.inrialpes.fr/people/mairal/"
                      moz-do-not-send="true">Julien Mairal</a> (INRIA)<br>
                  </li>
                  <li><a
href="https://www.iit.it/people/massimiliano-pontil"
                      moz-do-not-send="true">Massimiliano Pontil</a>
                    (IIT and University College London)<br>
                  </li>
                  <li><a
href="https://www.maths.usyd.edu.au/ut/people?who=DX_Zhou&sms=y"
                      moz-do-not-send="true">Dingxuan Zhou</a>
                    (University of Sydney)<br>
                  </li>
                </ul>
              </div>
              <div><br>
              </div>
              <div><b><i>- Call for abstracts -</i></b></div>
              <div><br>
              </div>
              <div>The DEEPK 2024 program will include <b>oral and
                  poster sessions</b>. Interested participants are
                cordially invited to submit an <b>extended abstract
                  (max. 2 pages)</b> for their contribution.  Please
                prepare your extended abstract submission in LaTeX,
                according to the provided stylefile and submit it in pdf
                format (max. 2 pages). Further extended abstract
                information is given at <a
                  class="moz-txt-link-freetext"
href="https://www.esat.kuleuven.be/stadius/E/DEEPK2024/call_for_abstracts.php"
                  moz-do-not-send="true">https://www.esat.kuleuven.be/stadius/E/DEEPK2024/call_for_abstracts.php</a>
                .</div>
              <div><br>
              </div>
              <div><b><i>- Schedule - </i></b><br>
              </div>
              <div>
                <ul>
                  <li><b>Deadline extended abstract submission:</b><br>
                    <b>Feb 8, 2024 </b></li>
                  <li>Notification of acceptance and presentation format
                    (oral/poster):<br>
                    Feb 22, 2024 </li>
                  <li>Deadline for registration:<br>
                    Feb 29, 2024 <br>
                  </li>
                  <li><b>International Workshop DEEPK 2024:</b><br>
                    <span style="color:#990000;font-weight:bold"> </span><span
                      style="font-weight: bold;">March 7-8, 2024</span>
                  </li>
                </ul>
              </div>
              <div><b><i>- Organizing committee - </i></b><br>
              </div>
              <div><br>
              </div>
              <div>Johan Suykens (Chair), Alex Lambert, Panos Patrinos,
                Qinghua Tao, Francesco Tonin</div>
              <div><br>
              </div>
              <div><b><i>- Other info -</i></b></div>
              <div><br>
              </div>
              <div>Please consult the DEEPK 2024 website <a
                  class="moz-txt-link-freetext"
href="https://www.esat.kuleuven.be/stadius/E/DEEPK2024"
                  moz-do-not-send="true">https://www.esat.kuleuven.be/stadius/E/DEEPK2024</a>
                for info on program, registration, location and venue.
                The event is co-sponsored by ERC Advanced Grant
                E-DUALITY and KU Leuven.<br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <div><br>
              </div>
              <br>
            </div>
          </div>
        </div>
      </div>
    </div>
  </body>
</html>