[SeminarioDeProbabilidad] Hoy seminario de PPE

[Liliana Peralta Hernández]

Estimada Comunidad,
Les  escribimos para invitarl en s a nuestra sesión del día de hoy 20 de marzo
del Seminario de Probabilidad y Procesos Estocásticos. El seminario se
llevara a cabo a las 17 hrs., en el salón S-105 del Departamento de
Matemáticas de la Facultad de Ciencias.

Ponente: Sameer Deshpande

Título: Beyond axis-alignment: Realizing the power of Bayesian Additive
Regression Trees in General Spaces

Resumen:
Default implementations of Bayesian Additive Regression Trees (BART)
represent categorical predictors using several binary indicators, one for
each level of each categorical predictor. Regression trees built with these
indicators partition the levels using a “remove one a time strategy.”
Unfortunately, an overwhelming majority of partitions of the levels cannot
be built with this strategy, severely limiting BART's ability to “borrow
strength” across groups of levels. We overcome this limitation with a new
class of regression trees built around decision rules based on linear
combinations of these indicators. Motivated by spatial applications with
areal data, we introduce a further decision rule prior that partitions the
areas into spatially contiguous regions by deleting edges from random
spanning trees of a suitably defined network. We implemented our new
regression tree priors in the flexBART package, which, compared to existing
implementations, often yields improved out-of-sample predictive performance
without much additional computational burden. We will conclude by
describing how the flexBART implementation can be further extended to fit
BART models over much more general input spaces. A preprint describing this
work is available at https://arxiv.org/abs/2211.04459.


L en s esperamos

Laura, Saraí, Liliana
------------ próxima parte ------------
Se ha borrado un adjunto en formato HTML...
URL: <http://higgs.matem.unam.mx/pipermail/seminarioprobabilidadyprocesos/attachments/20240320/f4024999/attachment.htm>