Proceedings of the VLDB Endowment · 2022 · 20 citations · 21 references
ComputingComputer ScienceData MiningMachine LearningNatural Language ProcessingOptimization For Machine LearningProbabilistic Machine LearningQuery OptimizationQuery ProcessingRepresentation LearningConditional Independence AssumptionIndependenceOptimizer DesignPredicate PushdownProxy Model ConstructionQuery ComplexityQuery Performance OptimizationUser-defined Functions
We consider accelerating machine learning (ML) inference queries on unstructured datasets. Expensive operators such as feature extractors and classifiers are deployed as user-defined functions (UDFs), which are not penetrable with classic query optimization techniques such as predicate push-down. Recent optimization schemes (e.g., Probabilistic Predicates or PP) assume independence among the query predicates, build a proxy model for each predicate offline, and rewrite a new query by injecting these cheap proxy models in the front of the expensive ML UDFs. In such a manner, unlikely inputs that do not satisfy query predicates are filtered early to bypass the ML UDFs. We show that enforcing the independence assumption in this context may result in sub-optimal plans. In this paper, we propose CORE, a query optimizer that better exploits the predicate correlations and accelerates ML inference queries. Our solution builds the proxy models online for a new query and leverages a branch-and-bound search process to reduce the building costs. Results on three real-world text, image and video datasets show that CORE improves the query throughput by up to 63% compared to PP and up to 80% compared to running the queries as it is.
21
Table of Integrals, Series, and Products
D. S., I. S. Gradshteyn, I.M. RYZHIK · Mathematics of Computation · 1966 · 8.2K citations
The Stanford CoreNLP Natural Language Processing Toolkit
Christopher D. Manning, Mihai Surdeanu, John Bauer et al. · 2014 · 7.2K citations · Full text
DeepPose: Human Pose Estimation via Deep Neural Networks
Alexander Toshev, Christian Szegedy · 2014 · 3.3K citations · Full text
Training linear SVMs in linear time
Thorsten Joachims · 2006 · 2K citations