• Publications
  • Influence
Polyhedral parallel code generation for CUDA
TLDR
This article addresses the compilation of a sequential program for parallel execution on a modern GPU. Expand
  • 279
  • 54
  • PDF
Tensor Comprehensions: Framework-Agnostic High-Performance Machine Learning Abstractions
TLDR
We demonstrate an end-to-end flow from the high-level Tensor Comprehensions (TC) language down to automatically generated kernels on GPUs. Expand
  • 198
  • 27
  • PDF
Wavelet methods in numerical analysis
TLDR
This chapter explains basic examples of wavelet methods in numerical analysis and investigates the way they can be generalized to multivariate functions. Expand
  • 206
  • 23
The Polyhedral Model Is More Widely Applicable Than You Think
TLDR
The polyhedral model is a powerful framework for automatic optimization and parallelization. Expand
  • 203
  • 14
  • PDF
Iterative optimization in the polyhedral model: part ii, multidimensional time
TLDR
We introduce a genetic algorithm with specialized operators that leverage the polyhedral representation of program dependences to traverse huge optimization spaces, achieving good performance improvements on large loop nests. Expand
  • 179
  • 14
  • PDF
Semi-Automatic Composition of Loop Transformations for Deep Parallelism and Memory Hierarchies
TLDR
We address this challenge by working on the program representation itself, using a semi-automatic optimization approach to demonstrate that current compilers suffer from unnecessary constraints and intricacies that can be avoided in a semantically richer transformation framework. Expand
  • 234
  • 13
  • PDF
Quasilinear subdivision schemes with applications to ENO interpolation
Abstract We analyze the convergence and smoothness of certain class of nonlinear subdivision schemes. We study the stability properties of these schemes and apply this analysis to the specific classExpand
  • 115
  • 13
Iterative Optimization in the Polyhedral Model: Part I, One-Dimensional Time
TLDR
We define a systematic exploration method to enumerate the space of all legal, distinct transformations in the class of loop transformation which can be expressed as one-dimensional affine schedules. Expand
  • 126
  • 13
  • PDF
Putting Polyhedral Loop Transformations to Work
TLDR
We seek to extend the scope and efficiency of iterative compilation techniques by searching not only for program transformation parameters but for the most appropriate transformations themselves. Expand
  • 147
  • 9
  • PDF
Wavelets and Multiscale Signal Processing
Multi-resolution analysis: The continuous point of view The discrete point of view The multivariate case. Wavelets and conjugate quadrature filters: The general case The finite case Wavelets withExpand
  • 216
  • 9
  • PDF
...
1
2
3
4
5
...