<html><head><style type='text/css'>p { margin: 0; }</style></head><body><div style='font-family: times new roman,new york,times,serif; font-size: 12pt; color: #000000'>Prezados colegas e alunos,<br><br>Esta semana mensagem é para lembra-los que esta semana a palestra do ciclo de Seminários de
Probabilidade e <span id="DWT79" class="ZmSearchResult">Sistemas</span> <span id="DWT1459" class="ZmSearchResult"><span id="DWT81" class="ZmSearchResult"><span id="DWT504" class="ZmSearchResult"><span id="DWT305" class="ZmSearchResult"><span id="DWT405" class="ZmSearchResult"><span id="DWT737" class="ZmSearchResult"><span id="DWT638" class="ZmSearchResult"><span id="DWT437" class="ZmSearchResult"><span id="DWT1554" class="ZmSearchResult"><span id="DWT446" class="ZmSearchResult">Complexos</span></span></span></span></span></span></span></span></span></span>,
ICMC-USP & UFSCar, São Carlos, será realizada na <span class="Object" id="OBJ_PREFIX_DWT447_com_zimbra_date">sexta</span> 19/06 na UFSCar. Uma
lista das palestras passadas, e as próximas, pode ser encontrada no
seguinte link: <br><br><span class="Object" id="OBJ_PREFIX_DWT82_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT505_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT306_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT406_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT738_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT639_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT438_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT1556_com_zimbra_url"><span class="Object" id="OBJ_PREFIX_DWT448_com_zimbra_url"><a target="_blank" href="http://www.icmc.usp.br/pessoas/pablor/seminar/pcs_seminar.html">http://www.icmc.usp.br/pessoas/pablor/seminar/pcs_seminar.html</a></span></span></span></span></span></span></span></span></span><br><br>Segue a informação da <span class="Object" id="OBJ_PREFIX_DWT83_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT506_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT307_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT407_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT739_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT640_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT439_com_zimbra_date"><span class="Object" id="OBJ_PREFIX_DWT1557_com_zimbra_date">palestra</span></span></span></span></span></span></span></span>.<br><h3>Nonparametric conditional density estimation in high dimensions: some theoretical aspects</h3>
<b>Speaker: Rafael Izbicki (UFSCar)</b>
<br>
Date: <span class="Object" id="OBJ_PREFIX_DWT449_com_zimbra_date">19/06/2015</span> • Time: 16h00 • Room: Sala de Seminários DEs-UFSCar<br><br>
<b>Abstract:</b> In some applications (e.g., in cosmology and
economics), the regression E[Z|x] is not adequate to represent the
association between a predictor x and a response Z because of
multi-modality and asymmetry of f(z|x); using the full density instead
of a single-point estimate can then lead to less bias in subsequent
analysis. However, there are currently no effective ways of estimating
f(z|x) when x represents high-dimensional, complex data. Here we propose
a fully nonparametric approach to conditional density estimation (CDE)
that reformulates CDE as a non-parametric orthogonal series problem
where the expansion coefficients are estimated by regression. By taking
such an approach, one can efficiently estimate conditional densities in
high dimensions by drawing upon the success in high-dimensional
regression. Depending on the choice of regression procedure, our method
can adapt to a variety of challenging high-dimensional settings with,
for example, a large number of irrelevant components or nonlinear
manifold structure in the data. We study the theoretical and empirical
performance of our proposed method, and we compare our approach with
traditional conditional density estimators on real-world as well as
simulated data.
<br><br>Estão todos convidados! Favor divulgar a possíveis interessados.<br><br>Um abraço,<br>Pablo<br></div></body></html>