Homepage for Eirik Fossgaard

About myself:

I was a PhD Research Fellow in applied mathematics at Department of Mathematics and Statistics, University of Tromsø, from 2000 to 2004. My PhD-project was supported by the Norwegian Research Council under BeMatA project no. 135971 and the University of Tromsø. My project advisor was professor Tor Flå.

Papers:

• Fast numerical algorithms for applying separable integral operators in higher dimensions.
  Eirik Fossgaard, Luca Frediani, Tor Flå and Kenneth Ruud (2007). Submitted.


• Method for optimal placement of sensors in a wired pipe drillstring .
  Eirik Fossgaard, Gerhard Nygaard and Erlend H. Vefring, IRIS Research (2007).
  Proceedings IADC/SPE Managed Pressure Drilling and Underbalanced Operations, Galveston, TX, 2007. IADC/SPE 108345.


• Parallelization of the integral equation formulation of the polarizable continuum model for higher-order response functions.
  Lara Ferrighi, Luca Frediani, Eirik Fossgaard, Kenneth Ruud (2006). Journal of Chemical Physics, 2006;125(15).


• Superlinear scaling in master-slave quantum chemical calculations using in-core storage of two-electron integrals.
  Eirik Fossgaard and Kenneth Ruud (2005). Journal of Computational Chemistry, 2006;27:326-333.


• An invariant bayesian model selection principle for gaussian data in a sparse representation. Eirik Fossgaard and Tor Flå (2004).
  IEEE Transactions on Information Theory, 2006;52(8) (pdf format).


• Alternative local discriminant bases using empirical expectation and variance estimation. Eirik Fossgaard (2000).
  Journal of Computational and Graphical Statistics, 2001;10(2) (pdf format).

Contact information:

• efossgaard_gmail.com

Projects:

• I worked as postdoc at Tromsø University Chemistry Department during: 01.05.2005 - 31.08.2006. The postdoc-project concerned development and implementation of fast and precise numerical algorithms using multiwavelets and numerical techniques for separation of variables on multi-dimensional operators for applying the Hamiltonian operator on multiparticle systems in quantum chemistry. This work was a joint project with fellow post.doc Luca Frediani. For details and results concerning this project, see paper: "Fast numerical algorithms for applying separable integral operators in higher dimensions", E. Fossgaard, L. Frediani, T. Flå and K. Ruud.

What I currently do:

• Geophysicist at StatoilHydro ASA. Location: Trondheim, Norway.

What I have been doing in the past:

• Research Scientist at IRIS-Petroleum Research. Location: Bergen, Norway.


• Post.Doc at The Chemistry Department at the University of Tromsø. (english version available)


• Senior Engineer at The Computer Center at the University of Tromsø. (english version available)


• Teacher and later a Research Fellow at The Mathematics Department at the University of Tromsø. (english version available)

Old stuff:

My research interests:

• Applications of multiresolution analysis / harmonic analysis / wavelets.


• Numerical operator calculus.


• Numerical analysis.


• Mathematical and statistical modeling / code length principles.

PhD-thesis and MSc-dissertation:

• Abstract of PhD thesis.(html format).


• Eirik Fossgaard (2004). An invariant bayesian model selection principle for gaussian data
  in a sparse representation. PhD thesis, University of Tromsø, 2004. (pdf format).


• Eirik Fossgaard (1997). Fast computational algorithms for the discrete wavelet transform
  and applications of localized orthonormal bases in signal classification. MSc dissertation, University of Tromsø, 1997. (pdf format).

Code and numerical data:

• Eirik Fossgaard (2004). The code (Ansi C) developed to do the numerical experiments
  reported in my PhD thesis. (gzipped tar archive).


• Eirik Fossgaard (2004). The Gauss Legendre quadrature nodes and weights of orders:
  1-33 and 48, 64, 96 to high precision: 34 decimal digits. (ASCII text file).


• Eirik Fossgaard (2004). The Legendre and interpolating multiwavelet filters of orders 1-16.
  (see: Alpert, Beylkin, Gines, Vozovoi: "Adaptive Solution of Partial Differential Equations in
  Multiwavelet Bases"), appoximate precision: (34 - N) decimal digits
  where N is the order of the multiwavelet. (gzipped tar archive).


• Eirik Fossgaard (2004). The code (Ansi C) used to generate the Legendre and interpolating
  multiwavelet filters and the Gauss/Legendre quadrature nodes and weights above
  (see: Alpert, Beylkin, Gines, Vozovoi: "Adaptive Solution of Partial Differential Equations
  in Multiwavelet Bases"). (gzipped tar archive).


• If you need more precision on the quadratures/multiwavelet filters (I did) you can rewrite the code provided above to use the MPFR-routines for addition/multiplication/division, or ask me, or possibly get them from MATLAB.