<div dir="ltr"><div class="gmail_quote"><div dir="ltr"><div class="gmail_quote"><div class="gmail_attr"><br></div><div dir="ltr"><p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif"">Prezados colegas, </p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif"">Nosso próximo encontro do COLMEA terá lugar na PUC-Rio, no
próximo dia 15. Será o primeiro encontro presencial depois de fevereiro de
2020.  Na ocasião teremos a seguinte
programação:</p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US">14:00 h</span></b><span lang="EN-US"> - <b>Paulo
Orenstein (IMPA)</b></span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US">Split conformal prediction and its extensions
to non-exchangeable data</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US">15:20 h - Ivailo Hartarsky (Université
Paris-Dauphine)</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US">Universality in bootstrap percolation and
kinetically constrained models</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;line-height:115%;color:black;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">Todos são muito bem-vindos.  </span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b>Local do evento: </b>Sala
de reuniões do Decanato do CTC</p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif"">12 º andar do prédio Cardeal Leme </p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif"">PUC-Rio, Gávea</p>

<p class="MsoNormal" style="line-height:12.65pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 10pt;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Mais informações
sobre o COLMEA podem ser encontradas através da homepage </span><a href="http://www.im.ufrj.br/~coloquiomea/" target="_blank"><span style="font-size:12pt">http://www.im.ufrj.br/~coloquiomea/</span></a><span style="font-size:12pt;color:black"></span></p>

<p class="MsoNormal" style="line-height:12.65pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 10pt;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Os resumos das
palestras estão no final da mensagem.  Agradecemos se puder divulgar. Em
anexo o cartaz de divulgação.  </span></p>

<p class="MsoNormal" style="line-height:12.65pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;margin:0cm 0cm 10pt;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Atenciosamente,</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">O comitê organizador: </span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Americo Cunha (UERJ)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Augusto Q. Teixeira (IMPA)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Evaldo M. F. Curado (CBPF)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">João Batista M. Pereira (UFRJ)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Leandro P. R. Pimentel (UFRJ)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Maria Eulalia Vares (UFRJ)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Nuno Crokidakis (UFF)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;color:black">Simon Griffiths (PUC-Rio)</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;line-height:115%"> </span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt;line-height:115%">--------------</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span style="font-size:12pt;line-height:115%">Resumos das palestras</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US" style="font-size:12pt;line-height:115%">Split
conformal prediction and its extensions to non-exchangeable data</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US">Paulo Orenstein (IMPA)</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US" style="font-size:12pt;line-height:115%">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<b> </b>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></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US" style="font-size:12pt;line-height:115%">This is joint work with Roberto Imbuzeiro Oliveira,
Thiago Ramos and João Vitor Romano.</span></p><p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US" style="font-size:12pt;line-height:115%"><br></span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US" style="font-size:12pt;line-height:115%">Universality
in bootstrap percolation and kinetically constrained models</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b><span lang="EN-US" style="font-size:12pt;line-height:115%">Ivailo
Hartarsky (Université Paris-Dauphine)</span></b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US" style="font-size:12pt;line-height:115%">The paradigmatic 2-neighbour bootstrap percolation
model is the following cellular automaton. Given a set of infected sites in Z^d,
we iteratively infect each site with at least 2 infected neighbours, while
infections never heal. We are then interested in whether and when the origin
becomes infected under this dynamics starting from an i.i.d. Bernoulli initial
infection. There is a naturally associated stochastic non-monotone model: the
Fredrickson-Andersen 2-spin facilitated one, in which the state of each site is
resampled to a Bernoulli variable at rate 1, provided it has at least 2
infected neighbours.</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US" style="font-size:12pt;line-height:115%">Of course, many related models have been considered,
replacing the 2-neighbour constraint by an increasing local
translation-invariant constraint (e.g. both the North and East neighbours are
infected). In this talk we will overview recent universality results for this
class of bootstrap percolation and its non-monotone stochastic counterpart
called kinetically constrained models. The outcome is a classification of all
rules in terms of the scaling of the infection time of the origin when the
density of infections approaches a possibly degenerate critical value.</span></p>

<p class="MsoNormal" style="margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,"sans-serif""><span style="font-size:12pt"> </span></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b> </b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><b> </b></p>

<p class="MsoNormal" style="margin:0cm 0cm 10pt;line-height:115%;font-size:11pt;font-family:Calibri,"sans-serif""><span lang="EN-US"> </span></p></div>
</div></div>
</div></div>