<!DOCTYPE html>
<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p class="MsoNormal" align="center"
      style="margin-bottom:16.0pt;text-align:center;
line-height:normal"><span
style="font-size:15.0pt;font-family:"Arial",sans-serif;
mso-fareast-font-family:"Times New Roman";color:black;mso-themecolor:text1;
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB">DeepMath
        2025: Call for Abstracts</span><span
style="font-family:"Times New Roman",serif;
mso-fareast-font-family:"Times New Roman";color:black;mso-themecolor:text1;
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"></span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">We are pleased to announce <b>DeepMath 2025</b>:
        the 7th installment of the Conference for the Mathematical
        Theory of Deep Learning (<a class="moz-txt-link-abbreviated"
          href="http://www.deepmath-conference.com">www.deepmath-conference.com</a>).
        This year DeepMath will be held at the <b>University of
          Michigan in Ann Arbor Michigan</b>. We have a stellar lineup
        of invited speakers including: </span><span
style="font-family:"Times New Roman",serif;
mso-fareast-font-family:"Times New Roman";mso-font-kerning:0pt;mso-ligatures:
none;mso-fareast-language:EN-GB"></span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"> </span></p>
    <ul style="margin-top:0cm" type="disc">
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Laura Balzano (University of Michigan)</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Rama Chellappa (Johns Hopkins University)</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Jelena Diakonikolas (University of Wisconsin)</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Surbhi Goel (University of Pennsylvania)</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Thomas Icard (Stanford University)</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l0 level1 lfo1;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Robert Nowak (University of Wisconsin)</span></li>
    </ul>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"> </span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">We are inviting researchers to present a
        poster during our poster session. Please submit your work no
        later than <b>August 31, 2025</b> on OpenReview using the
        following link:       </span><span
style="font-family:
"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";mso-font-kerning:
0pt;mso-ligatures:none;mso-fareast-language:EN-GB"></span></p>
    <p class="MsoNormal"
      style="margin-bottom:0cm;text-indent:36.0pt;line-height:
normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;
mso-fareast-font-family:"Times New Roman";color:black;mso-font-kerning:0pt;
mso-ligatures:none;mso-fareast-language:EN-GB">             </span><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"></span></p>
    <p class="MsoNormal"
      style="margin-bottom:0cm;text-indent:36.0pt;line-height:
normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:
"Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
EN-GB"><a
href="https://openreview.net/group?id=deepmath-conference.com/DeepMath/2025/Conference"
          moz-do-not-send="true"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;color:#1155CC">https://openreview.net/group?id=deepmath-conference.com/DeepMath/2025/Conference</span></a></span></p>
    <p class="MsoNormal"
      style="margin-bottom:0cm;text-indent:36.0pt;line-height:
normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:
"Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
EN-GB"> </span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">DeepMath is a highly interdisciplinary
        conference focused on understanding fundamental theory driving
        the success of Deep Learning. A principal goal of this
        conference is to bring together theoreticians working on deep
        learning from various disciplines and perspectives.  We,
        therefore, encourage submissions from researchers from diverse
        disciplines including but not limited to</span><span
style="font-family:
"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";mso-font-kerning:
0pt;mso-ligatures:none;mso-fareast-language:EN-GB"></span></p>
    <ul style="margin-top:0cm" type="disc">
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">​​Statistics</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Physics</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Computer science</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Neuroscience</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Mathematics</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Psychology</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l1 level1 lfo2;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Engineering​</span></li>
    </ul>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"> </span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">Topics may address any area of deep learning
        research such as:</span><span
style="font-family:"Times New Roman",serif;
mso-fareast-font-family:"Times New Roman";mso-font-kerning:0pt;mso-ligatures:
none;mso-fareast-language:EN-GB"></span></p>
    <ul style="margin-top:0cm" type="disc">
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Expressivity</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Generalization</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Optimization</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Representations</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Computation</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Network architectures​</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Recurrent networks</span></li>
      <li class="MsoNormal"
style="color:black;margin-bottom:0cm;line-height:normal;
     mso-list:l2 level1 lfo3;tab-stops:list 36.0pt;vertical-align:baseline"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
     "Times New Roman";mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:
     EN-GB">Emergent abilities like in-context learning, reasoning</span></li>
    </ul>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"> </span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">To complement the many conferences with
        applications and theory the focus for DeepMath will be
        exclusively on the theoretical and mechanistic understanding of
        the underlying properties of neural networks.</span></p>
    <p class="MsoNormal" style="margin-bottom:0cm;line-height:normal"><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB">Abstracts will not be made public (i.e., no
        official proceedings), and will be doubly-blind reviewed and
        selected for quality. All poster submissions should be properly
        anonymized in order to allow for blind refereeing. <b>Submissions
          should be no more than 1 page</b> although a second page may
        be used for references. Authors should submit a pdf file
        prepared using the <b>Latex style file available </b></span><span
style="font-family:"Times New Roman",serif;mso-fareast-font-family:"Times New Roman";
mso-font-kerning:0pt;mso-ligatures:none;mso-fareast-language:EN-GB"><a
href="https://storage.googleapis.com/wzukusers/user-23451351/documents/5c90d93c12862PuReXwq/DeepMath_template.zip"><b><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;color:#1155CC">here</span></b></a></span><span
style="font-size:10.0pt;font-family:"Arial",sans-serif;mso-fareast-font-family:
"Times New Roman";color:black;mso-font-kerning:0pt;mso-ligatures:none;
mso-fareast-language:EN-GB"> and should adopt all formatting, subject
        headings, font sizes, etc. defined therein. Submissions that
        fail to meet the format requirements will not be reviewed. The
        first author listed on the abstract is considered to be the
        presenting author. Each presenting author may submit only one
        abstract.</span><span
style="font-family:"Times New Roman",serif;
mso-fareast-font-family:"Times New Roman";mso-font-kerning:0pt;mso-ligatures:
none;mso-fareast-language:EN-GB"></span></p>
    <p class="MsoNormal"> </p>
    <p> </p>
    <p><br>
    </p>
    <div class="moz-forward-container"><br>
      <br>
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <p> </p>
      <br>
    </div>
  </body>
</html>