<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><blockquote type="cite" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0); -webkit-text-size-adjust: auto;"><div dir="ltr"><div class="WordSection1" style="page: WordSection1;"><div align="center"><div class="x-apple-transform-wrapper" style="height: 611.6072607260726px;"><table class="MsoNormalTable" border="0" cellspacing="0" cellpadding="0" width="600" style="transform-origin: 0px 0px; transform: scale(0.6171617161716172); width: 600px !important; height: 990.59375px !important;"><tbody><tr><td valign="top" style="background-color: white; padding: 0cm;"><div align="center"><table class="MsoNormalTable" border="0" cellspacing="0" cellpadding="0" width="600" style="width: 450pt;"><tbody><tr><td style="padding: 0cm;"><table class="MsoNormalTable" border="0" cellspacing="0" cellpadding="0" width="100%" style="width: 600px;"><tbody><tr><td style="padding: 0cm;"><div align="center"><table class="MsoNormalTable" border="0" cellspacing="0" cellpadding="0" width="540" style="width: 405pt;"><tbody><tr style="height: 22.5pt;"><td style="padding: 0cm; height: 22.5pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><b><span lang="EN-US" style="font-size: 18pt; font-family: Verdana, sans-serif; color: rgb(192, 0, 38);">Data Science<o:p></o:p></span></b></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><span style="font-size: 12pt; font-family: Arial, sans-serif;"> <o:p></o:p></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><span lang="EN-US" style="font-family: Verdana, sans-serif; color: rgb(192, 0, 38);">Academic Seminar – Hybrid</span><span style="font-size: 12pt; line-height: 16.799999237060547px;"><o:p></o:p></span></p></td></tr><tr style="height: 22.5pt;"><td style="padding: 0cm; height: 22.5pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><span lang="EN-US" style="font-size: 12pt; font-family: Arial, sans-serif;"> </span><span style="font-size: 12pt; font-family: Arial, sans-serif;"><o:p></o:p></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; text-align: justify;"><b><span lang="EN-US" style="font-size: 11.5pt; font-family: Verdana, sans-serif; color: rgb(51, 51, 51);">Title: </span></b><span style="font-size: 10.5pt; font-family: Verdana, sans-serif; color: black;">Split conformal prediction and its extensions to non-exchangeable data.<o:p></o:p></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><b><span lang="EN-US" style="font-size: 11.5pt; font-family: Verdana, sans-serif; color: rgb(51, 51, 51);">Speaker:</span></b><span lang="EN-US" style="font-size: 12pt; font-family: Arial, sans-serif;"> </span><span lang="EN-US" style="font-family: Arial, sans-serif;"><a href="https://w3.impa.br/~pauloo/" style="color: rgb(149, 79, 114);">Paulo Orenstein</a></span><span lang="EN-US" style="font-size: 12pt; font-family: Arial, sans-serif;"><br></span><b><span lang="EN-US" style="font-size: 11.5pt; font-family: Verdana, sans-serif; color: rgb(51, 51, 51);">University:</span></b><span lang="EN-US" style="font-size: 12pt; font-family: Arial, sans-serif;"> </span><u><span style="font-family: Verdana, sans-serif;"><a href="https://impa.br/" style="color: rgb(149, 79, 114);">IMPA</a></span></u><u><span style="font-size: 12pt; font-family: Arial, sans-serif;"><o:p></o:p></span></u></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; text-align: justify;"><b><span lang="EN-US" style="font-size: 11.5pt; font-family: Verdana, sans-serif; color: rgb(51, 51, 51);"> </span></b></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; text-align: justify; line-height: 16.866666793823242px;"><b><span lang="EN-US" style="font-size: 11.5pt; line-height: 17.633333206176758px; font-family: Verdana, sans-serif; color: rgb(51, 51, 51);">Abstract:</span></b><span lang="EN-US" style="font-family: Verdana, sans-serif; color: gray;"> </span><span lang="EN-US" style="font-size: 12pt; line-height: 18.399999618530273px; font-family: Arial, sans-serif;"> </span><span style="font-size: 10.5pt; line-height: 16.100000381469727px; font-family: Verdana, sans-serif; color: black;">Machine learning algorithms offer state-of-the-art predictive performance in a variety of domains, but often lack an associated measure of uncertainty quantification. Split conformal prediction is a leading tool to obtain predictive intervals with virtually no assumptions beyond data exchangeability. This crucial assumption, however, hinders its applicability to many important data, such as time series and spatially dependent processes. In this talk, we will introduce split CP and show how it can be extended to non-exchangeable settings through a small coverage penalty. The proposed framework, based on concentration of measure inequalities, works more generally than traditional split CP, and experiments corroborate our coverage guarantees even under highly dependent data.</span><span style="font-size: 12pt; line-height: 18.399999618530273px; font-family: Arial, sans-serif;"><o:p></o:p></span></p></td></tr><tr style="height: 22.5pt;"><td style="padding: 0cm; height: 22.5pt;"></td></tr><tr style="height: 114.1pt;"><td style="padding: 0cm; height: 114.1pt;"><table class="MsoNormalTable" border="0" cellspacing="0" cellpadding="0" width="430" style="width: 322.7pt;"><tbody><tr style="height: 29.1pt;"><td width="39" style="width: 28.95pt; padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><img border="0" width="21" height="22" id="_x0000_i1083" alt="image004.jpg" src="cid:18407031-EEF2-4667-8FDE-8E15A96BF9E6" style="width: 0.2187in; height: 0.2291in;"><span style="font-size: 12pt; line-height: 16.799999237060547px;"><o:p></o:p></span></p></td><td width="392" style="width: 293.75pt; padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><span style="font-size: 10.5pt; line-height: 14.699999809265137px; font-family: Verdana, sans-serif; color: black;"><span dir="ltr" style="text-decoration: underline;">May 30, 2022</span></span><span style="font-size: 10.5pt; line-height: 14.699999809265137px; font-family: Verdana, sans-serif;"> - </span><b><span style="font-size: 10pt; line-height: 14px; font-family: Verdana, sans-serif; color: black;">Excepcionalmente segunda-feira</span></b><o:p></o:p></p></td></tr><tr style="height: 29.1pt;"><td style="padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><img border="0" width="21" height="21" id="_x0000_i1084" alt="image005.jpg" src="cid:7504CD0D-4486-40CF-8AFC-43F3B19F49E6" style="width: 0.2187in; height: 0.2187in;"><o:p></o:p></p></td><td style="padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><span style="font-size: 10.5pt; line-height: 14.699999809265137px; font-family: Verdana, sans-serif; color: black;"><span dir="ltr" style="text-decoration: underline;">12:00 p.m.</span> de São Paulo, Brasil (UTC/GMT -<span dir="ltr" style="text-decoration: underline;">03:00</span>)</span><o:p></o:p></p></td></tr><tr style="height: 29.1pt;"><td style="padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><img border="0" width="21" height="21" id="_x0000_i1085" alt="image006.jpg" src="cid:5B9BDA5A-5E93-4112-96C0-CFB8ADF85BA1" style="width: 0.2187in; height: 0.2187in;"><o:p></o:p></p></td><td style="padding: 0cm; height: 29.1pt;"><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><o:p> </o:p></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><span style="font-size: 10.5pt; line-height: 14.699999809265137px; font-family: Verdana, sans-serif; color: black;">Paulo Renato de Souza room, 2nd floor - Building 1<o:p></o:p></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><b> </b></p><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif; line-height: 15.399999618530273px;"><b><span style="font-size: 10.5pt; line-height: 14.699999809265137px; font-family: Verdana, sans-serif; color: black;"><a href="https://zoom.us/j/91497126748" style="color: rgb(149, 79, 114);">Click here to join</a><o:p></o:p></span></b></p></td></tr></tbody></table><p class="MsoNormal" style="margin: 0cm 0cm 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;"><o:p> </o:p></p></td></tr></tbody></table></div></td></tr></tbody></table></td></tr></tbody></table></div></td></tr></tbody></table></div></div></div></div></blockquote><br><div dir="ltr">Sent from Hedibert's iPhone</div></body></html>