SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning
Summary: Reinforcement learning drives on-the-fly join ordering without statistics or cost models, achieving regret-bounded execution. Execution splits into time slices testing orders, merging results, and switching plans with a custom engine; experiments show gains vs MonetDB/Postgres with negligible overhead. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Immanuel Trummer (Cornell University)
- 2. Junxiong Wang (Cornell University)
- 3. Deepak Maram (Cornell University)
- 4. Samuel Moseley (Cornell University)
- 5. Saehan Jo (Cornell University)
- 6. Joseph Antonakakis (Cornell University)
BibTeX Citation
@inproceedings{trummer_sigmod19,
title = {{SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning}},
author = {Trummer, Immanuel and Wang, Junxiong and Maram, Deepak and Moseley, Samuel and Jo, Saehan and Antonakakis, Joseph},
series = {{SIGMOD} '19},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3299869.3300088},
url = {https://dl.acm.org/doi/10.1145/3299869.3300088},
year = {2019}
}
Incoming Citations (Sorted by Pagerank)
Showing 39 of 39 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 22 of 22 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next