The Tensor Algebra Compiler

Linear algebra is a work-horse of numerical computing. Tensor algebra is a generalization of linear algebra with applications in scientific computing, machine learning, and data analytics. 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 on dense and sparse tensors in different formats. The combinations are infinite, which makes it impossible to manually implement and optimize them all. I will present the first compiler technique to automatically generate kernels for any compound linear and 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. For more information, see tensor-compiler.org.

Date:
Speakers:
Fredrik Kjolstad 
Affiliation:
MIT