Sparsetir: Composable abstractions for sparse compilation in deep learning
Sparse tensors are rapidly becoming critical components of modern deep learning
workloads. However, developing high-performance sparse operators can be difficult and …
workloads. However, developing high-performance sparse operators can be difficult and …
The sparse abstract machine
We propose the Sparse Abstract Machine (SAM), an abstract machine model for targeting
sparse tensor algebra to reconfigurable and fixed-function spatial dataflow accelerators …
sparse tensor algebra to reconfigurable and fixed-function spatial dataflow accelerators …
Teaal: A declarative framework for modeling sparse tensor accelerators
Over the past few years, the explosion in sparse tensor algebra workloads has led to a
corresponding rise in domain-specific accelerators to service them. Due to the irregularity …
corresponding rise in domain-specific accelerators to service them. Due to the irregularity …
Mosaic: An interoperable compiler for tensor algebra
We introduce Mosaic, a sparse tensor algebra compiler that can bind tensor expressions to
external functions of other tensor algebra libraries and compilers. Users can extend Mosaic …
external functions of other tensor algebra libraries and compilers. Users can extend Mosaic …
Indexed Streams: A Formal Intermediate Representation for Fused Contraction Programs
S Kovach, P Kolichala, T Gu, F Kjolstad - Proceedings of the ACM on …, 2023 - dl.acm.org
We introduce indexed streams, a formal operational model and intermediate representation
that describes the fused execution of a contraction language that encompasses both sparse …
that describes the fused execution of a contraction language that encompasses both sparse …
The EDGE language: Extended general einsums for graph algorithms
In this work, we propose a unified abstraction for graph algorithms: the Extended General
Einsums language, or EDGE. The EDGE language expresses graph algorithms in the …
Einsums language, or EDGE. The EDGE language expresses graph algorithms in the …
Looplets: A language for structured coiteration
W Ahrens, D Donenfeld, F Kjolstad… - Proceedings of the 21st …, 2023 - dl.acm.org
Real world arrays often contain underlying structure, such as sparsity, runs of repeated
values, or symmetry. Specializing for structure yields significant speedups. But automatically …
values, or symmetry. Specializing for structure yields significant speedups. But automatically …
Minimum cost loop nests for contraction of a sparse tensor with a tensor network
R Kanakagiri, E Solomonik - Proceedings of the 36th ACM Symposium …, 2024 - dl.acm.org
Sparse tensor decomposition and completion are common in numerous applications,
ranging from machine learning to computational quantum chemistry. Typically, the main …
ranging from machine learning to computational quantum chemistry. Typically, the main …
Finch: Sparse and Structured Array Programming with Control Flow
From FORTRAN to NumPy, arrays have revolutionized how we express computation.
However, arrays in these, and almost all prominent systems, can only handle dense …
However, arrays in these, and almost all prominent systems, can only handle dense …
Compilation of Modular and General Sparse Workspaces
Recent years have seen considerable work on compiling sparse tensor algebra
expressions. This paper addresses a shortcoming in that work, namely how to generate …
expressions. This paper addresses a shortcoming in that work, namely how to generate …