<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    [Apologies if you receive multiple copies of this CFP]<br>
    <br>
    <p>Dear all, <br>
    </p>
    <p>ICPRAI conference has delayed the submission date of its paper to
      the <u>15th of january 2022</u>,  this also applies for the
      special session "BGMV-XAI 2022 : Vis&ML for XAI - Bridging the
      Gap between ML and Visualization communities for eXplainable
      Artificial Intelligence".
    </p>
    Please find enclosed the updated CFP.
    <br>
    <br>
    Best regards
    <br>
    <br>
    Romain Bourqui, Romain Giot, Wojciech Samek
    <br>
    <br>
    ---
    <br>
    <br>
    <div class="moz-text-html" lang="x-unicode">
      <header id="title-block-header">
        <h1 class="title">Vis&ML for XAI - Special Session of ICPRAI
          2022<br>
        </h1>
        <h1 class="title"><font size="2">Bridging the Gap between ML and
            Visualization communities for eXplainable Artificial
            Intelligence</font></h1>
        <p class="subtitle"><br>
        </p>
      </header>
      <table width="100%" cellspacing="0" cellpadding="0" border="0">
        <tbody>
          <tr>
            <td valign="middle" align="center"><a
                href="https://bgmv-xai.labri.fr/"><img id="header_logo"
                  style="margin-right: 15px;"
                  src="https://bgmv-xai.labri.fr/imgs/paris_rectangle_400_200.jpg"
                  alt="" name="header_logo"></a></td>
            <td valign="middle" align="left">
              <div id="header_title"><a style="color: #000000;"
                  href="https://icprai2022.sciencesconf.org/">ICPRAI
                  2022 - 3rd International Conference on Pattern
                  Recognition and Artificial Intelligence<br>
                  <br>
                </a></div>
              <div id="header_subtitle">June 1-<sup> </sup>3, 2022<br>
                Doctoral consortium: May 31, 2022</div>
              <div> </div>
              <div id="header_subtitle"><a
                  href="https://icprai2022.sciencesconf.org/">Paris
                  (France)</a></div>
            </td>
          </tr>
        </tbody>
      </table>
      <h1 id="call-for-paper">Call for paper</h1>
      <h2 id="about">About</h2>
      <p>The rise of machine learning approaches, and in particular deep
        learning, has led to a significant increase in the performance
        of AI systems. However, it has also raised the question of the
        reliability and explicability of their predictions for
        decision-making (<em>i.e.</em>, the black-box issue of the deep
        models). Such shortcomings also raise many ethical and political
        concerns that prevent wider adoption of this potentially highly
        beneficial technology, especially in critical areas, such as
        healthcare, self-driving cars or security. It is therefore
        critical to understand how their predictions correlate with
        information perception and expert decision-making. The objective
        of eXplainable AI (XAI) is to open this black-box by proposing
        methods to understand and explain how these systems produce
        their decisions.</p>
      <p>Research work in XAI is currently carried out in parallel by
        the Machine Learning and the Information Visualization
        communities using methodologies and competencies from their own
        field. This special session hosted by the <a
          href="https://icprai2022.sciencesconf.org/">ICPRAI conference</a>,
        endorsed by IAPR, is an opportunity to fill the gap between
        Machine Learning and Information Visualization communities and
        to promote new joint research paths.</p>
      <h2 id="topics">Topics</h2>
      <p>Here are the main, but not limited to, topics of interest:</p>
      <ul>
        <li>Trust, Uncertainty, Fairness, Accountability and
          Transparency</li>
        <li>Explainable/Interpretable Machine Learning</li>
        <li>Information visualization for models or their predictions</li>
        <li>Interactive applications for XAI</li>
        <li>XAI Evaluation and Benchmarks</li>
        <li>Human-AI interface and interaction design</li>
        <li>Sample-centric and Dataset-centric explanations</li>
        <li>Attention mechanisms for XAI</li>
        <li>Pruning with XAI</li>
      </ul>
      <p>We expect papers written by researchers from both communities,
        with a preference for works that imply a joint research (e.g.,
        visualization experts with machine learning experts). Paper
        selection will be achieved by a program committee of experts in
        Machine Learning and experts in Information Visualization;
        additionally, each paper will be reviewed by at least one expert
        of the two communities.</p>
      <h1 id="paper-submission">Paper submission</h1>
      <p>A paper can be submitted via the EasyChair online submission
        system at the following address: <a
          href="https://easychair.org/my/conference?conf=icprai2022"
          class="uri moz-txt-link-freetext">https://easychair.org/my/conference?conf=icprai2022</a>.
        You must select the item <em>ICPRAI 2022 - SS - Vis&ML for
          XAI: Bridging the gap between machine learning and
          visualization communities for eXplainable Artificial
          Intelligence</em>.</p>
      <br>
      <h1 id="important-dates">Important dates</h1>
      <ul>
        <li><b><code>15/01/2022</code></b><b>: Submission deadline
            (revised deadline)</b><br>
        </li>
        <li><code>08/03/2022</code>: Author notification</li>
        <li><code>22/03/2022</code>: Camera ready deadline</li>
        <li><code>01/06/2022--03/06/2022</code>: ICPRAI 2022 - 3rd
          International Conference on Pattern Recognition and Artificial
          Intelligence</li>
      </ul>
    </div>
    <br>
    More information is available at: <a class="moz-txt-link-freetext"
      href="https://bgmv-xai.labri.fr/">https://bgmv-xai.labri.fr/</a>
  </body>
</html>