Compiler Support for Sparse Tensor Computations in MLIR

  • Bik, Aart J. C.
  • Koanantakool, Penporn
  • Shpeisman, Tatiana
  • Vasilache, Nicolas
  • Zheng, Bixia
  • Kjolstad, Fredrik
Publication date
February 2022
Publisher
Association for Computing Machinery (ACM)
Language
English

Abstract

Sparse tensors arise in problems in science, engineering, machine learning, and data analytics. Programs that operate on such tensors can exploit sparsity to reduce storage requirements and computational time. Developing and maintaining sparse software by hand, however, is a complex and error-prone task. Therefore, we propose treating sparsity as a property of tensors, not a tedious implementation task, and letting a sparse compiler generate sparse code automatically from a sparsity-agnostic definition of the computation. This paper discusses integrating this idea into MLIR

Extracted data

Loading...

Related items

Automatic generation of efficient sparse tensor format conversion routines
  • Chou, Stephen
  • Kjolstad, Fredrik
  • Amarasinghe, Saman
November 2020

© 2020 Owner/Author. This paper shows how to generate code that efficiently converts sparse tensors ...

Sparse tensor algebra compilation
  • Kjølstad, Fredrik Berg.
January 2020

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Comp...

The Tensor Algebra Compiler
  • Kjolstad, Fredrik
  • Kamil, Shoaib
  • Chou, Stephen
  • Lugato, David
  • Amarasinghe, Saman
February 2017

Tensor and linear algebra is pervasive in data analytics and the physical sciences. Often the tensor...

We use cookies to provide a better user experience.