Computations in Science Seminars

Wednesdays at KPTC 206, unless otherwise specified

The Kersten Physics Teaching Center is on the corner of 57th Street and Ellis Avenue.

Discussion over bag-lunch at 12:15 PM. Talk starts at 12:30 PM.

INFORMATION FOR SPEAKERS

Upcoming seminars

Previous seminars

This seminar series is organized by David Biron, email address, Wendy Zhang, email address, and Leo Kadanoff, email address.

 A photo of M51 Galaxy

Whirlpool Galaxy M51
February 10, 2010
Steve Berry, University of Chicago
e-mail: email address
Exploring Landscapes in Many Dimensions: How We and Many-Atom Systems Do It
We can treat the behavior of many-atom systems in terms of how they move on an effective potential surface of 3N-6 independent variables and one dependent variable, the internal energy, for a system of N atoms. This is adequate for most dielectric systems; collections of metal atoms are more complex and sometimes require considering multiple potential surfaces. However, restricting ourselves to dielectrics, at least to get into the subject, we recognize that we face a problem of trying to understand what goes on in a space of very many dimensions, if we want to treat anything beyond the very simplest systems. We find ourselves confronting such questions as, "Why can some systems, cooled from their liquid state, invariably find their way to special, often well-ordered structures, while others 'get lost' in any of a large number of amorphous structures?" This is obviously related to the issue of protein folding, among others. Can we characterize such surfaces in terms of their stationary points? Can we use master equations to describe what happens on these surfaces, in terms of their well-to-well kinetics? How is the topography of a multidimensional potential related to the interparticle forces? These are some of the issues we face in trying to address a challenge of true complexity.
February 17, 2010
Juan Restrepo, University of Arizona
e-mail: email address, Faculty contact: Ridgway Scott, email address
Climate: When Data Fail Us, Nonlinear/Non-Gaussian Estimation
State estimation techniques are used in weather and climate prediction, hydrogeology, seismology, as a way to blend model output and real data in order to improve on predictions from the exclusive use of the model or the data alone. Techniques that are based upon least-squares ideas, such as the family of Kalman Filter/Smoothers, or Variational Data Assimilation, are optimal in linear/Gaussian problems. However, they often fail in problems in which nonlinearities are important and/or when Gaussianity in the statistics cannot be assumed. Even linearization may fail, and so do ensemble techniques that make nonlinear predictions but rely on linear analyses. These comprise the practical state of the art, at least in weather forecasting and in hydrogeology. I will describe these as well as how failures arise in these methods. We have created a number of nonlinear/non-Gaussian data assimilation techniques. Our present efforts are to make them computationally practical as well as to use of these to do problems that are otherwise intractable using conventional means. One such application is in Lagrangian data assimilation: here we tackle the problem of blending data that has been sampled along paths, which when blended in traditional ways on Eulerian grids will lead to loss of critical features even though the estimates may be variance-minimizing.
February 24, 2010
Mike Wilde, Argonne National Laboratory
e-mail: email address, Faculty contact: David Biron, email address
March 3, 2010
Daniel Rothman, Massachusetts Institute of Technology
e-mail: email address, Faculty contact: Leo Kadanoff, email address
March 10, 2010
Chris Myers, Cornell University
e-mail: email address, Faculty contact: Leo Kadanoff, email address
March 17, 2010 (during March meeting)
(open date)
March 24, 2010 (^)
(open date)
March 31, 2010 (^)
(open date)
April 6, 2010 (#) (JFI Seminar)
David Chandler, University of California, Berkeley
e-mail: email address, Faculty contact: Leo Kadanoff, email address
Structure of trajectory space, broken symmetry and a glass transition
Super-cooling a liquid often produces glass -- a solid with no apparent structural order. Unlike crystallization, a glass transition is not accompanied by a thermodynamic singularity. Nevertheless, a phase transition can underlie the formation of glass. Unlike equilibrium order-disorder phenomena, this transition appears as a singularity in a partition function of dynamical histories. I describe this transition -- its order parameters and phase diagrams.
April 7, 2010 (#)
David Chandler, University of California, Berkeley
e-mail: email address, Faculty contact: Leo Kadanoff, email address
Importance sampling of trajectory space: throwing ropes over rugged mountain passes, in the dark
The statistical physics of trajectory space has a distinguished history -- from Onsager's formulation of non-equilibrium thermodynamics, to Martin-Siggia-Rose theory, and so on. In recent years, it has provided principles that, among other things, facilitate computer simulations of rare events, and numerical studies of non-equilibrium phase transitions. I describe this development and its applications in a few illustrative cases.
April 14, 2010 (#)
Itai Cohen, Cornell University
e-mail: email address, Faculty contact: Wendy Zhang, email address
April 21, 2010 (#)
James Shapiro, The University of Chicago
e-mail: email address, Faculty contact: Wendy Zhang, email address
Evolution in the 21st Century
Ideas about evolution were first formulated before biology and genetics developed into the sophisticated sciences they are today. Molecular biology has had a profound impact on our understanding of how organisms are related to each other and how they change over time. Genome sequencing reveals the evolutionary record as it remains in the DNA of living organisms and constitutes a test of theories about how evolution has occurred. Discoveries about the molecular and cellular nature of evolutionary changes show that this remains a vital and exciting area of science with many new theoretical and experimental possibilities. In particular, genome sequences teach us that many key events in evolution have been accompanied by major and rapid changes in the content and organization of cell DNA that affected numerous characters at the same time. These kinds of changes were unknowable to the pioneers of evolutionary thinking and have not yet been included in conventional statements about how the evolutionary process operates. Incorporating lessons from the DNA record, recent observations of evolution in action, and experiments about the biological processes of genome change make it possible to formulate a 21st Century view that is consistent with other developments in the molecular life sciences.
April 28, 2010 (#)
(open date)
May 5, 2010 (^)
Daan Frenkel, Trinity College, UK
e-mail: email address, Faculty contact: Tom Witten, email address
May 12, 2010
Neil Turok, Perimeter Institute
e-mail: email address, Faculty contact: Leo Kadanoff, email address
May 19, 2010
Mark Goulian, University of Pennsylvania
e-mail: email address, Faculty contact: David Biron, email address
May 26, 2010
(open date)
June 2, 2010
Cristian Huepe, Northwestern University
e-mail: email address, Faculty contact: David Biron, email address
June 9, 2010
Priscilla C. Frisch, The University of Chicago
e-mail: email address, Faculty contact: David Biron, email address
June 16, 2010
(open date)
June 23, 2010
(open date)
June 30, 2010
(open date)
July 7, 2010
(open date)
July 14, 2010
(open date)
July 21, 2010
(open date)
July 28, 2010
(open date)
August 4, 2010
(open date)
August 11, 2010
(open date)

(&) : When Wendy Zhang is unavailable for the seminar.

(^) : When Leo Kadanoff is unavailable for the seminar.

(#) : When David Biron is unavailable for the seminar.

Back to JFI home page.

Valid XHTML 1.0 Transitional