BOWLING GREEN STATE UNIVERSITY
          DEPARTMENT OF MATHEMATICS AND STATISTICS CALENDAR
 
                         Week of February 17-21

Tuesday, February 18

 2:30 PM  STATISTICS SEMINAR  - Room 459 MSC
          P. K. Sen, Distinguished Lukacs Professor, BGSU. 
          "Rank tests for the location model"

 3:15 PM  Coffee
 3:45 PM  COLLOQUIUM  - Room 459 MSC
          Sergey Shpectorov, Ohio State University
          "Almost classical geometries and sporadic groups"
          Abstract: Old and new results concerning geometries of
            sporadic simple groups will be discussed, with the
            emphasis on the similarities and connections with the
            classical Tits' theory of buildings.

Wednesday, February 19
      
 3:30 PM  STATISTICS AND PROBABILITY SEMINAR  - Room 447 MSC
          Jim Albert, Dept. of Mathematics and Statistics, BGSU. 
          "Hierarchical modeling and Bayes factors"
          Abstract: I will introduce Bayesian hierarchical model in
            the setting of learning about proportions of successful
            heart transplant operations for a number of hospitals.
            A Poisson/gamma model is used to fit the data.  We first
            describe how one checks this model using classical methods
            and then describe alternative Bayesian methods for model
            checking.

Thursday, February 20

 2:30 PM  SCIENTIFIC COMPUTATION SEMINAR  - Room 459 MSC
          J. Gordon Wade, Dept. of Mathematics and Statistics, BGSU.
          "Necessary conditions for total variation minimization and
           minimal surfaces"
          Abstract: We consider numerical approaches for solving
            problems involving total variation minimization for image
            reconstruction and for minimal surfaces.  The problems are
            formulated as nonquadratic minimization problems for image
            reconstruction. The first-order necessary condition or
            "Euler equation" for a minimizer yields a quasilinear
            elliptic equation of the form

                          [ L^*L + A(u) ] u = -L^*z

            with boundary conditions. Here, L is a bounded linear
            operator and A(u) is a standard self-adjoint second order
            elliptic operator in which the coefficient a depends on u,
            by

                   [a(u)](x) = c/sqrt(|grad u(x)|^2 + b^2)

            where b and c are small constants. A common and effective
            strategy for solving the Euler equation is the fixed point
            method.

            Total variation minimization has been quite beneficial in
            recent applications in image denoising and deblurring,
            laser interferometry, electrical tomography, and
            estimation of permeabilities in porous media flow
            models. Its main advantage is that it improves the
            conditioning of the optimization problem while not
            penalizing discontinuities in the reconstructed image. The
            main difficulty in its use lies in the fact that the Euler
            equation is nonlinear with rapidly varying coefficients
            and can have a rather large number (e.g., 640-squared) of
            degrees of freedom.

 3:30 PM  ALGEBRA SEMINAR  - Room 459 MSC
          Curtis Bennett, Dept. of Mathematics and Statistics, BGSU. 
          "Connected components of the Universal Twin"

Friday, February 21

 3:15 PM  Coffee
 3:45 PM  COLLOQUIUM  - Room 459 MSC
          Keith Kearns, University of Arkansas
          "Minimal varieties of abelian algebras"
          Abstract: A variety is an equationally defined class of
            algebras.  A variety is minimal if it is minimal under
            inclusion among varieties that contain nontrivial
            algebras.  This talk is about recent results concerning
            minimal varieties, including the classification of locally
            finite, abelian, minimal varieties.

This announcement and a schedule of future colloquia are available on
the Worldwide Web; see http://www.bgsu.edu/departments/math/.
If you would like to place a link to this calendar on your page,
use html code 
<A HREF="http://www-math.bgsu.edu/~zirbel/weekly_cal.html">
Department of Mathematics and Statistics Calendar</A>
If you wish to be placed on the e-mail distribution list, or have 
comments or material for the calendar, send email to