Lefschetz Center for Dynamical Systems Seminar
Brown University Center for Statistical Sciences Seminar
Department of Biostatistics, Bioinformatics and Epidemiology | |
Abstract: This talk addresses modeling the joint outcomes of a recurrent event process and an associated longitudinal biomarker. Examples where such data may arise include those monitoring repeated heart attacks and related markers such as cholesterol levels or inflammation measures, or studies assessing association between markers for diabetes control and recurrent oral complications. We describe a latent class model in which the event process and longitudinal outcome are conditionally independent given the latent class. The model for the recurrent event process accommodates the effect of accumulating event occurrences on the subject, as well as the effects of interventions performed after each event occurrence. We illustrate the behavior of this joint model and discuss estimation and inference.
This is joint work with Professor Edsel Pena of the University of South Carolina.
**Sponsored by the Charles P. Sisson II Memorial Lectureship
***Co Sponsored by The Marshall Woods Lectureships Foundation
of Fine Arts
Center for Fluid Mechanics Seminar
And
The Fluids, Thermal and Chemical Processes Group
Of
The Division of Engineering
Seminar Series
Department of Chemical and Biological Engineering | |
Abstract: Organization and evolution in systems having many degrees of freedom and operating away from thermodynamic equilibrium ("dissipative systems") is of fundamental interest for its relevance to life and for potential technological applications. Interestingly, such systems can be designed rationally based on a set of heuristic rules. My talk will focus on the implementation of these rules to engineer two classes of dynamic, self-organizing systems: one based on hydrodynamic vortex-vortex interactions, and the other on electrostatic forces mediated by contact electrification. I will illustrate how ensembles of identical components interacting by simple potentials can be "complexified" and ultimately developed into self-organizing machines, microfluidic devices and detection systems."
Brown Applied Mathematics Pattern Theory and Vision Seminar
Abstract: Motion segmentation is a rich source of training data for learning to segment objects by their static image properties. Background subtraction can distinguish between moving objects and their backgrounds, and the techniques of statistical machine learning can be used to capture information about objects' shape, size, color, brightness, and texture properties. Presented with a new, static image, the trained model can infer the proper segmentation of the objects present in a scene. The algorithm presented in this work uses the techniques of Markov random field modeling and belief propagation inference, and outperforms a standard segmentation algorithm on the object segmentation task.
Brown Analysis Seminar
Applied Mathematics Colloquium
THANKSGIVING TURKEY TALK
Abstract: This past summer, a young student attending a math camp for bright kids asked me "How do we know when a problem is solved?" This seemingly innocent question has deep implications within the methodology and philosophy of mathematics. Some of these will be explored in the talk.
PDE Seminar
<--- 2004 Index