On Efficient Approximate Queries over Machine Learning Models

Summary: Framework for approximate queries over ML predictions that minimizes expensive oracle (human/DNN) calls by combining cheap proxy scores with selective oracle sampling for precision- and recall-target queries. Two regimes—Proxy Quality (PQA/PQE) and Core Set Closure (CSC/CSE)—offer provable guarantees and empirically outperform prior work. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 13531
Venue: VLDB
Year: 2023
Pagerank: 5.2829539e-05
Overall Rank: 9,353 | 35.84%
DOI: 10.14778/3574245.3574273

Incoming Non-self Citations Over Time

Authors

1. Dujian Ding (University of British Columbia)
2. Sihem Amer-Yahia (CNRS; University Grenoble Alpes)
3. Laks Lakshmanan (University of British Columbia)

BibTeX Citation

@article{ding_vldb23,
        title = {{On Efficient Approximate Queries over Machine Learning Models}},
        author = {Ding, Dujian and Amer-Yahia, Sihem and Lakshmanan, Laks},
        journal = {PVLDB},
        series = {{VLDB} '23},
        volume = {16},
        number = {4},
        pages = {918--931},
        doi = {10.14778/3574245.3574273},
        url = {https://doi.org/10.14778/3574245.3574273},
        year = {2023}
}

Incoming Citations (Sorted by Pagerank)

Showing 3 of 3 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
8,829	Cut Costs, Not Accuracy: LLM-Powered Data Processing with Guarantees	2026	SIGMOD	5.3613784e-05
10,476	On Efficient Approximate Aggregate Nearest Neighbor Queries over Learned Representations	2026	SIGMOD	5.093636e-05
10,795	Scalable Complex Event Processing on Video Streams	2025	SIGMOD	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 8 of 8 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
295	Accelerating Machine Learning Inference with Probabilistic Predicates	2018	SIGMOD	0.00022238183
1,519	Top-k Query Evaluation with Probabilistic Guarantees	2004	VLDB	0.00010513777
2,293	Extending Relational Query Processing with ML Inference	2020	CIDR	8.7949378e-05
2,898	Approximate Selection with Guarantees using Proxies	2020	VLDB	7.978725e-05
4,114	Optimizing Machine Learning Inference Queries with Correlative Proxy Models	2022	VLDB	6.8941194e-05
5,827	Top-K Deep Video Analytics: A Probabilistic Approach	2021	SIGMOD	6.0744172e-05
6,761	Efficiently Answering Durability Prediction Queries	2021	SIGMOD	5.7822533e-05
8,848	A Generalized Approach for Reducing Expensive Distance Calls for A Broad Class of Proximity Problems	2021	SIGMOD	5.3577837e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,184	100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models: [Experiments & Analysis]	2026	SIGMOD
2	295	Accelerating Machine Learning Inference with Probabilistic Predicates	2018	SIGMOD
3	4,136	Approximating Predicates and Expressive Queries on Probabilistic Databases	2008	PODS
4	4,789	Learned Approximate Query Processing: Make it Light, Accurate and Fast	2021	CIDR
5	9,882	Demonstration of Accelerating Machine Learning Inference Queries with Correlative Proxy Models	2022	VLDB
6	4,114	Optimizing Machine Learning Inference Queries with Correlative Proxy Models	2022	VLDB
7	11,624	Accelerating Queries over Unstructured Data with ML	2021	CIDR
8	3,907	Accelerating Approximate Aggregation Queries with Expensive Predicates	2021	VLDB
9	11,845	Query-Driven Learning for Next Generation Predictive Modeling & Analytics	2019	SIGMOD
10	50	Efficient Query Evaluation on Probabilistic Databases	2004	VLDB