Lefschetz Center for Dynamical Systems Seminar
Abstract: Estimation of transport and mixing for coherent structures in geophysical flow models requires solving initial-and-boundary problems for the flow field first. In practice, initial and boundary data are obtained from incomplete and noisy flow-field observations. In this study we focus on mid-latitude eddies that play an important role in the oceans' heat, mass and momentum fluxes. We assimilate very sparse data to estimate flow dynamics and the associated transports by designing a data-adaptive observing system that helps improve estimation's efficiency and accuracy. To do so, data assimilation methodology and dynamical systems theory are systematically combined. Using the extended Kalman filter of sequential estimation theory, we first demonstrate how subsurface flow information can be extracted efficiently from the available surface observations. Chaotic mixing techniques are then applied to estimate the flux carried by the eddies and the fluid's mixing with the surrounding flow field.
Stochastic Systems Seminar
Abstract:
The setting is that
is a stationary sequence taking values
in a subset of Euclidean space. The objective is to construct
prediction algorithms
such that in
some standard sense,
where either t is fixed and n -->
or
n is fixed and t -->
.
Process models and mixing conditions are not postulated.
Research during the past few years has yielded some weakly and strongly consistent prediction methods as well as considerable understanding of conditions under which consistent prediction is possible and when it is not. This fundamental problem area has been related to basic issues in information theory, gambling, and large-deviation theory.
This talk will survey results and implications of the theory, with emphasis on contributions by the speaker and his collaborators.
Brown University Center for Statistical Sciences Seminar
Abstract: It is becoming increasingly common for epidemiologists to consider randomizing intact social units (e.g. families, schools, communities) rather than independent individuals in experimental trials. Reasons are diverse, but include administrative convenience, a desire to reduce the effect of treatment contamination and the need to avoid ethical issues which might otherwise arise. Dependencies among cluster members typical of such designs must be considered when determining sample size and analysing the resulting data. Failure to adjust standard statistical methods for within-cluster dependencies may result in severely underpowered studies and in spuriously elevated Type I error rates. The purpose of this talk is to review the key issues in the design and analysis of cluster randomization trials. These ideas will be illustrated using data from several recently completed studies.
Special Department of Mathematics Colloquium
Brown Applied Mathematics Pattern Theory and Vision Seminar
Brown Analysis Seminar
Division of Applied Mathematics, Special Seminar
A Live Demonstration Using the SP2 at Brown There will be demonstrations at a time to be announced. | |
If you would like to talk with the speaker, please call Professor George Karniadakis at extension 1217. |
Abstract: Deep Blue, the IBM Chess Machine, recently made history by becoming the first computer to beat the human World Chess Champion, Garry Kasparov, in a regulation game. Although Kasparov came back to win the 6-game match, the IBM Deep Blue system demonstrated that a sophisticated chess system can be developed using the IBM RS6000/SP parallel processor, which meets a long standing challenge in computer science. In this talk we will describe the architecture of the Deep Blue system, illustrate the strong and weak points of Deep Blue thru highlights of the match. We will also discuss the implications of the event on chess and technology.
PDE/Lefschetz Center Seminar
Department of Mathematics Colloquium
<--- 1997 Index