A Tensor Algebra Compiler for Sparse Differentiation

Sparse tensors are prevalent in many data-intensive applications. However, existing automatic differentiation (AD) frameworks are tailored towards dense tensors, which makes it a challenge to efficiently compute gradients through sparse tensor operations. This is due to irregular sparsity patterns t...

Full description

Bibliographic Details

Published in:	Proceedings / International Symposium on Code Generation and Optimization pp. 1 - 12
Main Authors:	Shaikhha, Amir, Huot, Mathieu, Hashemian, Shideh
Format:	Conference Proceeding
Language:	English
Published:	IEEE 02.03.2024
Subjects:	Automatic Differentiation Codes Dictionaries DSL Optimization Pipelines Scalability Semi-Ring Dictionaries Sparse Tensor Algebra Tensors
ISSN:	2643-2838
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!