Tensor algebra is a powerful tool with applications in machine learning, data analytics, engineering and the physical sciences. Tensors are often sparse and compound operations must frequently be computed in a single kernel for performance and to save memory. Programmers are left to write kernels for every operation of interest, with different mixes of dense and sparse tensors in different formats. The combinations are infinite, which makes it impossible to manually implement and optimize them all. This paper introduces the first compiler technique to automatically generate kernels for any compound tensor algebra operation on dense and sparse tensors. The technique is implemented in a C++ library called taco. Its performance is competitive with best-in-class hand-optimized kernels in popular libraries, while supporting far more tensor operations.
Thu 26 OctDisplayed time zone: Tijuana, Baja California change
10:30 - 12:00 | Optimizing Compilation and VerificationOOPSLA at Regency C Chair(s): Gregor Richards University of Waterloo | ||
10:30 22mTalk | The Tensor Algebra Compiler OOPSLA Fredrik Kjolstad MIT CSAIL, Shoaib Kamil Adobe, Stephen Chou MIT CSAIL, David Lugato CEA, France, Saman Amarasinghe MIT DOI | ||
10:52 22mTalk | TreeFuser: A Framework for Analyzing and Fusing General Recursive Tree Traversals OOPSLA Laith Sakka Purdue University, Kirshanthan Sundararajah Purdue University, Milind Kulkarni Purdue University DOI | ||
11:15 22mTalk | Verifying Spatial Properties of Array Computations OOPSLA Dominic Orchard University of Kent, UK, Mistral Contrastin , Matthew Danish University of Cambridge, UK, Andrew Rice University of Cambridge, UK DOI | ||
11:37 22mTalk | GLORE: Generalized Loop Redundancy Elimination upon LER-Notation OOPSLA DOI | ||