The  Weizmann  Institute  of  Science
                  Faculty of Mathematics and Computer Science


                            Computer Science Seminar

                                  Achi Brandt 



                                 will speak on


                       Complexity of Scientific Computing

Abstract:
How many computer operations are required to calculate the mass of an
elementary particle from quantum chromodynamics theory? How many operations are
needed to calculate the electronic structure of a molecule, or of some
condensed matter, from the equations of quantum mechanics? Or to find the
folding and function of a protein? Or the properties of a bacterium? What
should be the computational complexity in calculating a large number of
eigenfunctions of a given differential operator? How expensive should the
simulation of a turbulent flow be?  What is the computational cost of
 designing an airplane? Or a drug? How fast and accurately can the internal
 makeup and physiology of a body be reconstructed from various medical-imaging
data?

In most of these tasks, several types of complexity factors multiply each
other, resulting in formidable computational bottlenecks; just faster computer
would not help. New hierarchical approaches for organizing computation in the
physical space can eliminate or drastically reduce each of these complexity
factors. The resulting overall cost now tends to be nearly {\em linear} in the
amount of {\em distinct} local patterns appearing in the problem, which is
often many orders-of-magnitude smaller than the number of degrees of freedom.






                      The lecture will take place in the 
                     Lecture Hall, Room 1, Ziskind Building
                           on Monday, April 17, 2000
                                    at 14:30