Center for Statistical Sciences Seminar Series
Associate Professor of Statistics, Psychology and Psychiatry, University of Illinois at Urbana-Champaign | |
Refreshments beginning at 3:15pm |
Abstract: Recent technological progress has given scientists a tremendous opportunity to observe different brain imaging modalities, each of which reflect different dimensions of the fascinating and highly complex brain processes. This talk will cover some of the interesting and challenging statistical problems in analyzing brain signals.
The first part of the talk will be an overview of a coherent and unified body of statistical signal processing methods that are based on the SLEX (Smooth Localized complex EXponentials) library. The basic building blocks in this approach are the SLEX waveforms which are essentially time-localized and orthonormal generalizations of the Fourier waveforms. The SLEX library provides a systematic and natural way of representing non-stationary signals, estimating the time-dependent spectra and identifying transient spectral and cross-spectral features that are useful for statistical learning applications. The SLEX methods are able to handle massive data sets because they utilize computationally efficient algorithms. Moreover, as a matter of practical importance, the SLEX methods give results that are easy to understand because they are time-dependent analogues of the classical Fourier methods for stationary signals. Finally, under the SLEX models, we develop theoretical results of consistency for spectral estimation and signal classification. An application of the SLEX methods to EEGs and MEGs will be presented.
In the second part of the talk, I will outline my current research program whose theme is an integrated time-frequency-spatial analysis of signals collected in an experimental design. I will develop wavelet-based functional mixed effects models for the time-varying spectra whose fixed and random subject-specific components use wavelet representations. The algorithm for estimating the fixed function, covariate effects and between-subject variation is computationally efficient and easy to implement. Moreover, the resulting estimates are spatially-adaptive with the ability to preserve sharp peaks and localized features. Finally, the talk will conclude with contemporary issues in brain signal analysis and the potential contributions of Statistics towards advancing our understanding of brain activity.
Department of Physics Colloquia
Brown Analysis Seminar
Mathematics Department Colloquium>
Abstract: Recent advances in the theory of quasilinear and fully nonlinear elliptic equations will be presented based on nonlinear potential theory and harmonic analysis methods. A solution to the existence problem in the viscosity sense, together with a complete characterization of removable singularities, will be discussed for a class of model equations which involve the k-Hessian and p-Laplacian operators. Our main tools include dyadic models, Wolff's potential estimates, and weak continuity theorems. This work is joint with Nguyen Cong Phuc.
Brown University
Joint Materials/Solid Mechanics Seminar Series
Department of Applied Physics and Applied Mathematics Materials Research Science and Engineering Center Columbia University, New York, New York | |
Mechanical, and Electrical Properties | |
Abstract: Electric fields are useful in combining nanocomponents to form nanomaterials in a spatially selective manner. The assembly of thin films of CdSe nanocrystals by electrophoretic deposition is described, along with the mechanical and optical properties of these films. The nature of the assembly and film properties are quite dependent on the nanocrystal charge and capping ligands. The precisely controlled dielectrophoretic placement of carbon nanotubes in electrode geometries is also presented.
2006 SPDE Workshop on
Advances and Challenges in the
Solution of Stochastic Partial Differential Equation
Sponsored by AFOSR
8:30-9:00 Registration/Coffee/Juice Each days sessions begin at 9:00 a.m.
| |
A. Chorin, University of California, Berkeley, B. Rozovsky, Brown University, G. Papanicolau, Stanford University, P. Kotelenez, Case Western Reserve University, G. Karniadakis, Brown University, S. V. Lototsky, University of South Carolina, J. Mattingly, Duke University, Saturday, October 21, 2006 -- Speakers: B. Oksendal, University of Oslo, H. J. Kushner, Brown University, A. Stuart, Warwick University, UK, J. Tribbia, National Center for Atmospheric Research, R. Ghanem, University of South Carolina, T. Hou, CalTech, E. Vander-Eijden, Courant Institute, New York University, Sunday, October 22, 2006 -- Speakers: D. Crisan, Imperial, UK, J. Glimm, Stony Brook University and Brookhaven National Lab, B. Birnir, University of California, Santa Barbara, R. Temam, Indiana University |
<--- 2006 Index