DBScholar

Back to papers

EinDecomp: Decomposition of Declaratively-Specified Machine Learning and Numerical Computations for Parallel Execution

Summary: EinDecomp declaratively specifies tensor computations via extended Einstein notation and rewrites them as tensor-relational programs. It optimizes graph decompositions for intra-operator execution across GPUs/CPUs, generalizing data- and model-parallelism. (summarized by gpt-5.6-luna on Jul 24 2026)

Paper ID
14062
Venue
VLDB
Year
2025
Pagerank
5.2112879e-05
Overall Rank
9,834 | 32.54%
DOI
10.14778/3734839.3734858

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{bourgeois_vldb25,
        title = {{EinDecomp: Decomposition of Declaratively-Specified Machine Learning and Numerical Computations for Parallel Execution}},
        author = {Bourgeois, Daniel and Ding, Zhimin and Jankov, Dimitrije and Li, Jiehui and Sleem, Mahmoud and Tang, Yuxin and Yao, Jiawen and Yao, Xinyu and Jermaine, Chris},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {7},
        pages = {2240--2253},
        doi = {10.14778/3734839.3734858},
        url = {https://doi.org/10.14778/3734839.3734858},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank Citing Paper Year Venue Pagerank
8,513 Galley: Modern Query Optimization for Sparse Tensor Programs 2025 SIGMOD 5.4119882e-05
10,514 Automated Tensor-Relational Decomposition for Large-Scale Sparse Tensor Computation 2026 VLDB 5.093636e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 12 of 12 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers