Establishes learnability of signed-measure selectivity predictors with OOD generalization bounds and derives practical strategies that improve OOD accuracy and latency in query-driven models.
Proceedings of the VLDB Endowment 12 , 3 (2018), 210–222
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Practical Theory of Generalization in Selectivity Learning
Establishes learnability of signed-measure selectivity predictors with OOD generalization bounds and derives practical strategies that improve OOD accuracy and latency in query-driven models.