Learned Approximate Query Processing: Make it Light, Accurate and Fast
Summary: DBEst++: a lightweight learned AQP engine that blends word embeddings with compact neural regressors for joint density estimation and aggregation-value prediction, enabling many small models to cover broad analytical workloads. Robust to high-cardinality categoricals and updates; empirically outperforms learned and sampling-based AQP on TPC‑DS/Flights in accuracy, latency and memory. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Qingzhi Ma (University of Warwick)
- 2. Ali M. Shanghooshabad (University of Warwick)
- 3. Mehrdad Almasi (University of Warwick)
- 4. Meghdad Kurmanji (University of Warwick)
- 5. Peter Triantafillou (University of Warwick)
BibTeX Citation
@inproceedings{ma_cidr21,
address = {Amsterdam, Netherlands},
series = {{CIDR} '21},
title = {{Learned Approximate Query Processing: Make it Light, Accurate and Fast}},
booktitle = {Proceedings of the {Conference} on {Innovative} {Data} {Systems} {Research}},
author = {Ma, Qingzhi and Shanghooshabad, Ali M. and Almasi, Mehrdad and Kurmanji, Meghdad and Triantafillou, Peter},
year = {2021}
}
Incoming Citations (Sorted by Pagerank)
Showing 13 of 13 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 18 of 18 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 5,906 | Approximate Query Engines: Commercial Challenges and Research Opportunities | 2017 | SIGMOD |
| 2 | 6,206 | Combining Aggregation and Sampling (Nearly) Optimally for Approximate Query Processing | 2021 | SIGMOD |
| 3 | 10,185 | A Backend-Agnostic Compiler for Approximate Query Processing with Probabilistic Tensor Algebra | 2026 | SIGMOD |
| 4 | 323 | DeepDB: Learn from Data, not from Queries! | 2020 | VLDB |
| 5 | 10,634 | Efficient Approximate Query Processing with Block Sampling | 2025 | CIDR |
| 6 | 7,066 | Aero: Adaptive Query Processing of ML Queries | 2025 | SIGMOD |
| 7 | 3,366 | AQP++: Connecting Approximate Query Processing With Aggregate Precomputation for Interactive Analytics | 2018 | SIGMOD |
| 8 | 9,353 | On Efficient Approximate Queries over Machine Learning Models | 2023 | VLDB |
| 9 | 11,845 | Query-Driven Learning for Next Generation Predictive Modeling & Analytics | 2019 | SIGMOD |
| 10 | 1,799 | DBEst: Revisiting Approximate Query Processing Engines with Machine Learning Models | 2019 | SIGMOD |