Steven I Reeves

CSE , Computational Research Division

Contact Information

Steven I Reeves
MS 50A-3111
Lawrence Berkeley National Lab
1 Cyclotron Rd.
Berkeley, CA 94720

510-486-6900 (fax)

Affiliation and Research Interests

I am a Computer Systems Engineer in the Center for Computational Sciences and Engineering (CCSE) in the Computational Research Division of the Computing Sciences Directorate at Lawrence Berkeley National Laboratory. My research is in computation methods for variously physical systems as well as interpolation for image and signal processing. Notably, I utilize the tools from Gaussian Process modeling to probabilistically predict fine grid values from a coarse grid state variable (AMR). I am also researching the generation of super-resolution images from low resolution images utilizing GP interpolation and convolutional neural networks. Additionally I work in high performance computing (HPC) -- leveraging the newest generation Graphics Processing Units in distributed computing systems, as well as more traditional CPU based super-computers, for scientific applications. Previously, I have worked on problems in Nonlinear-Dynamical systems, utilizing HPC and numerical analysis.

Research Interests

  • Gaussian Process Modeling for Adaptive Mesh Refinement
  • GP + Convolutional Neural Networks for Image Processing
  • Developing a N-body/hydro/MHD cosmology code (see NYX)
  • General Processing with Graphics Processing Units
  • Machine Learning in Science
  • Numerical methods for PDEs and Nonlinear Dynamical Systems



P-L Buono, V. In, P. Longhini, L. Olender, A. Palacios and S.I. Reeves, "Phase Drift on Networks of Coupled Crystal Oscillators For Precision Timing". Physical Review E

P-L Buono, B. Chan, J. Ferreira, A. Palacios, S.I. Reeves, P. Longhini, V. In, "Symmetry-Breaking Bifurcations and Patterns of Oscillations in Rings of Crysatal Oscillators". SIAM Journal on Applied Dynamical Systems