RIT on Statistics Archives for Academic Year 2015


Inference from High Dimensional Models

When: Tue, September 2, 2014 - 3:30pm
Where: MTH 1313
Speaker: Organizational Meeting: Stat Faculty (UMCP) -


High-dimensional Linear Regression

When: Tue, September 16, 2014 - 3:30pm
Where: MTH 1313
Speaker: Tingni Sun (Dept. of Mathematics, UMCP) -
Abstract: I will review some important results for sparse linear regression in the high-dimensional setting. If time permits, I will discuss the inference about covariance matrices and their inverse.

Selected Problems for High-Dimensional Models

When: Tue, September 23, 2014 - 3:30pm
Where: MTH 1313
Speaker: Prof. Paul Smith (Department of Mathematics, UMCP) -


High Dimensional Data and Biological Applications

When: Tue, September 30, 2014 - 7:00am
Where: MATH 1313
Speaker: Ying Han (UMCP) -


Variable Screening in High Dimensional Space

When: Thu, October 16, 2014 - 3:30pm
Where: MTH 1313
Speaker: Jinhang Xue (UMCP) -


Classification in High Dimensional Models

When: Tue, October 21, 2014 - 3:30pm
Where: MTH 1313
Speaker: Paul Smith, Department of Mathematics, UMCP

Operator Analysis and Diffusion Based Embeddings for Heterogeneous Data Fusion

When: Tue, November 4, 2014 - 3:00pm
Where: MTH 2300
Speaker: Wojciech Czaja (UMCP) -
Abstract:
As new sensing modalities emerge and the presence of multiple sensors per platform becomes widespread, it is vital to develop new algorithms and techniques which can fuse this data. Many of previous attempts to deal with the problem of heterogeneous data integration for the applications in data classification were either highly data dependent or relied on imply fusing classifier outputs. In this talk we shall examine several related approaches: graph fusion, operator fusion, and feature space fusion. They are all associated with graph diffusion processes generated by appropriately designed operators. Our results do not make any assumptions about the data and can be easily extended to new additional modalities. We shall illustrate these concepts with an example of hyperspectral imagery and lidar data fusion.


Structural change on Sparsity

When: Tue, November 11, 2014 - 3:30pm
Where: MTH 1313
Speaker: Professor Yuan Liao ( (UMCP)) -
Abstract: In the sparse modeling literature, it has been crucially assumed that the identities of important regressors are invariant across the population and across the individuals in the collected sample. In practice, however, the sparsity structure may not be always invariant in the population. For instance, the identities of important genes may be different, depending on the environmental temperature. I allow a possible structural change regarding the identities of important regressors in the population. One of the strengths is that it does not require to know whether the structural change is present, or where it occurs. We also identify the location where the possible change point occurs.

Bayesian Methods for Big Data

When: Tue, November 18, 2014 - 3:30pm
Where: MTH 1313
Speaker: Chen Wang (UMCP) -


Three case studies in the field of transportation engineering/modeling

When: Tue, December 2, 2014 - 3:30pm
Where: MTH 1313
Speaker: Prof. Cinzia Cirillo (Dept. of Civil and Environmental Engineering, UMCP) -


Can big data help in the production of reliable local area statistics?

When: Tue, December 2, 2014 - 4:00pm
Where: MTH 1313
Speaker: Prof. Partha Lahiri ( Joint Program in Survey Methodology (JPSM)) -


Multiple Testing on High Dimensional Data

When: Tue, December 9, 2014 - 3:30pm
Where: MTH 1313
Speaker: Ling Cai

High Dimensional Inferences

When: Tue, February 10, 2015 - 3:30pm
Where: MATH 1313
Speaker: Prof. Yuan Liao (UMCP) -


RIT on High Dimensional Inference

When: Tue, March 3, 2015 - 3:30pm
Where: MATH 1313
Speaker: Qi Liu (UMCP) -


Sure Independence Screening for Ultrahigh Dimensional Feature Space" Author(s): Jiaqing Fan and Jinchi Lv

When: Tue, May 5, 2015 - 3:30pm
Where: MATH 1313
Speaker1: Jinhang Xue (UMCP) -
Speaker2: Cheng Wang (UMCP), Title: Inference in additively separable models with a high dimensional set of conditioning variables", the author is Damian Kozbur who is student of Christian Hansen.