Skip to content
NewA calmer, faster ZeroHour

When Plans Change Answers: Formalizing Cost-Accuracy Optimization for Semantic Queries

arXiv cs.AI / cs.LG / cs.CLby Kyoungmin Kim ()1 min read2 · 1 read
AI summary · grok-4.7Summaries by Vani AI

Paper formalizes cost-accuracy optimization for semantic queries whose learned predicates change answers with the plan.

The paper formalizes cost-accuracy optimization for semantic query engines, where machine-learned predicates make the query plan change both cost and answers. Calibrated decision confidence produces an expected error, and weighting by each decision's contribution estimates plan quality without labeled data. Accuracy is plan-invariant under pointwise-deterministic policies, selection pushdown is not quality-sound for self-calibrated escalation bands, and choosing which tuples to drop is NP-hard. Simulations use a synthetic workload; evaluation on real engines is left for future work.

Key facts
  • Formal semantics and plan-equivalence hierarchy for semantic operators.
  • Expected quality is polynomial-time under bag semantics.
  • Choosing which tuples to drop is NP-hard.
  • Only synthetic simulations; real-engine evaluation is future work.
Explain this AI news with Vani AIWhat was announced, what is new and why it matters, in a new chat

From the original article

1 min read at arxiv.org

In semantic query engines, predicates are evaluated by machine-learned models, and the choice of a query plan affects not only the cost of a query but also its result. Existing systems either apply a fixed threshold to each semantic operator or tune accuracy per operator, without accounting for how errors propagate through joins. We give a formal problem definition for cost-accuracy optimization of such queries.

arXiv cs.AI / cs.LG / cs.CL

By Kyoungmin Kim · arxiv.org

Read the full article at arXiv cs.AI / cs.LG / cs.CL

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.