<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<b style="font-weight:normal">
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">Criteo is organizing a competition of machine learning on aggregated, differentially private data in conjunction
with the AdKDD'21 workshop. We thought it could be of interest to you and to the community of researchers and practitioners interested in machine learning and privacy.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">In two words, the tasks are to learn click and sales prediction models that operate on aggregated, noisy data.
We believe that the aggregation (features + labels), the noise level, and the high cardinality of features present novel, unique challenges.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">The competition will last until July 31st and the top 3 winners of each task will share $20,000 of prize money
and be invited to present their solution to the workshop at KDD'21.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">Important information:</span></p>
<ul style="margin-top:0;margin-bottom:0;padding-inline-start:48px">
<li dir="ltr" style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span>Workshop website:
</span><a href="https://www.adkdd.org/"><span style="color:#1155cc;text-decoration:underline;text-decoration-skip-ink:none">https://www.adkdd.org/</span></a></p>
</li><li dir="ltr" style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span>Competition website:
</span><a href="https://competitions.codalab.org/competitions/31485"><span style="color:#1155cc;text-decoration:underline;text-decoration-skip-ink:none">https://competitions.codalab.org/competitions/31485</span></a></p>
</li><li dir="ltr" style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span>Contact address:
</span><a href="mailto:adkdd21challenge@googlegroups.com"><span style="color:#1155cc;text-decoration:underline;text-decoration-skip-ink:none">adkdd21challenge@googlegroups.com</span></a></p>
</li></ul>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">Data and evaluation code will be open-sourced at the end of the challenge to permit publication of original solutions
in relevant research venues.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:#000000;font-weight:400">Eustache Diemert, for the challenge organization team at Criteo</span></p>
</b></div>
</body>
</html>