Probabilistic Aggregation Query Processing with Selective Predicates

Home

Discover

Bookmarks

Keep Up With Top Trending Papers In CS / AI / ML ...

Source arXiv

Databases

Probabilistic Aggregation Query Processing with Selective Predicates

Accelerating Approximate Aggregation Queries with Expensive Predicates

We develop and analyze a query processing algorithm that leverages proxies (proxies) to accelerate aggregation queries over large unstructured datasets with selective predicates that are computed using expensive deep neural networks (dnns).

We show that proxies converges at an optimal rate in a novel analysis of stratified sampling with draws that may not satisfy the predicate.

We further show that the proposed query processing algorithm outperforms on baselines on six real-world datasets, reducing labeling costs by up to 2.3x.

Authors

Daniel Kang, John Guibas, Peter Bailis, Tatsunori Hashimoto, Yi Sun, Matei Zaharia

◐ Recommended Members

◐ Latest Activity