Philipp Witte
Georgia Institute of Technology
48 Papers
168 Citations
Philipp Witte is an academic researcher from Georgia Institute of Technology. The author has contributed to research in topics: Computer science & Compiler. The author has an hindex of 10, co-authored 36 publications. Previous affiliations of Philipp Witte include Imperial College London & University of British Columbia.
Chat about Author
Papers
Devito (v3.1.0): an embedded domain-specific language for finite differences and geophysical exploration
Mathias Louboutin,Michael Lange,Fabio Luporini,Navjot Kukreja,Philipp Witte,Felix J. Herrmann,Paulius Velesko,Gerard J. Gorman +7 more
TL;DR: Devito as mentioned in this paper is a domain-specific language for implementing high-performance finite-difference partial differential equation solvers for exploration of the Earth's subsurface using full waveform inversion and reverse-time migration.
Architecture and Performance of Devito, a System for Automated Stencil Computation
Fabio Luporini,Mathias Louboutin,Michael Lange,Navjot Kukreja,Philipp Witte,Jan Hückelheim,Charles R. Yount,Paul H. J. Kelly,Felix J. Herrmann,Gerard J. Gorman +9 more
TL;DR: The architecture of the Devito compiler, as well as the performance optimizations that are applied when generating code, are presented and the effectiveness of such performance optimizations is demonstrated using operators drawn from seismic imaging applications.
A large-scale framework for symbolic implementations of seismic inversion algorithms in Julia
Philipp Witte,Mathias Louboutin,Navjot Kukreja,Fabio Luporini,Michael Lange,Gerard J. Gorman,Felix J. Herrmann +6 more
TL;DR: Writing software packages for seismic inversion is a very challenging task, since problems such as full-waveform inversion or least-squares imaging are both algorithmically and computationally demanding due to the large number of unknown parameters and the fact that waves are propagated over many wavelengths.
69
•Posted Content
Architecture and performance of Devito, a system for automated stencil computation
Fabio Luporini,Michael Lange,Mathias Louboutin,Navjot Kukreja,Jan Hückelheim,Charles R. Yount,Philipp Witte,Paul H. J. Kelly,Felix J. Herrmann,Gerard J. Gorman +9 more
TL;DR: Devito as discussed by the authors is a framework capable of generating highly-optimized code given symbolic equations expressed in Python, specialized in affine (stencil) codes, which is performed by the Devito compiler through a series of intermediate representations.
53
Compressive least-squares migration with on-the-fly Fourier transforms
TL;DR: This work introduces an algorithm for low-cost sparsity-promoting least-squares migration using on the-fly Fourier transforms to formulate the least-Squares migration objective function.
41