ICIAM Maxwell Prize 2011

Image

Awarded to Vladimir Rokhlin (New Haven, USA).

Vladimir Rokhlin of Yale University has been selected for the 2011 ICIAM Maxwell Prize for his work on fast multipole methods which have revolutionized fields like numerical electromagnetism for radar and molecular dynamics for chemistry.

Vladimir Rokhlin is a professor in the Computer Science Department and the Mathematics Department of Yale University. He was born in Voronezh, Russia on 4 August 1952 and received his M.S. in Mathematics in 1973 at the Vilnius University in Lithuania. He earned his Ph.D. in Applied Mathematics at Rice University in 1983.

Vladimir Rokhlin has had a profound impact on scientific computing and applied mathematics, most notably in developing "analysis-based fast algorithms". These include the fast multipole method for the Laplace equation, the fast multipole method for the Helmholtz equation, and the non-equispaced fast Fourier transform and also most recently in randomized matrix compression schemes. He has also made fundamental contributions to inverse scattering and to approximation theory.

Rokhlin was the first person who took a systematic approach to combining approximation theory, the classical theory of special functions, and modern computer science to reduce the computational cost associated with handling the basic integral operators of mathematical physics. Earlier fast algorithms (like the fast Fourier transform) had had great impact, but they were brittle. They had required uniform data structures and could not cope with complex geometries. An interesting consequence of the approximate nature of this new class of methods is that they are more flexible as well as being more robust.

His work on Fast Multipole Methods (FMM) has been cited as one of the ten algorithmic revolutions of the second half of the 20th century. These methods have revolutionized fields like numerical electromagnetism for radars and molecular dynamics for chemistry because the computing time to solve the problems is drastically reduced. For instance, for an airplane described by ten thousand points the radar cross-section can be computed in forty thousand operations instead of the millions of billions by earlier methods. FMM depends heavily on mathematical analysis and proper computer implementation and here too Vladimir Rokhlin has had a major role. By his close contact as advisor to industries and his interests for applications and above all his mathematical genius he has also produced exceptional Ph.D. students.

Subcommittee for the 2011 ICIAM Maxwell Prize

  • Olivier Pironneau (U Paris VI, France), chair;
  • Jonathan Borwein (Newcastle, Australia);
  • Ingrid Daubechies (Princeton, USA);
  • Heinz Engl (Vienna, Austria);
  • Leslie Greengard (New York, USA);
  • Tim Pedley (Cambridge, UK);
  • Graeme Wake (Massey, NZ).