## Ciclo de Palestras 2014 – 2° Semestre

##### Palestras do Departamento de Metodos Estatísticos - Instituto de Matemática - UFRJ

**2º semestre de 2014**

As palestras ocorreram no Auditório do Laboratório de Sistemas Estocásticos (LSE), sala I-044b, as 15:30 h, a menos de algumas exceções devidamente indicadas.

###### 15/12 (excepcionalmente em uma 2a feira as 15:00hs na C116)

###### 26/11 (Colóquio Interinstitucional "Modelos Estocásticos e Aplicações" as 14:00 no CBPF)

In this seminar entanglement will be taken as a dynamic quantity on its own, that evolves due to the unavoidable interaction of the entangled system with its surroundings. I will introduce the main aspects of entanglement dynamics in open quantum systems, portraying its richness and complexity. After setting the stage, I will present two different approaches two deal with entanglement dynamics: First, for bipartite systems I'll present a deterministic dynamical equation for entanglement. Second, in order to cope with many-body systems, I'll resort to a statistical description of typical entanglement dynamics. The latter relies solely on geometrical aspects of the space of states.

###### 24/11 (excepcionalmente em uma 2a feira as 16:30 na C116)

###### 18/11 (execepcionalmente em uma 3a feira as 09:30 na C116)

The paper revisits Bayesian group lasso and uses spike and slab priors for group variable selection. In the process, the connection of our model with penalized regression is demonstrated, and the role of posterior median for thresholding is pointed out. We show that the posterior median estimator has the oracle property for group variable selection and estimation under orthogonal design while the usual group lasso has suboptimal asymptotic estimation rate when variable selection consistency is achieved. Next we consider bi-level selection problem and propose the Bayesian sparse group lasso again with spike and slab priors to select variables both at the group level and also within a group. We demonstrate via simulation that the posterior median estimator of our spike and slab models has excellent performance for both variable selection and estimation.