mlsanityData to 14 Oct 2026

One Model to Rule them All: Towards Zero-Shot Learning for Databases

arXiv 2105.00642 ↗First posted 3 May 2021cs.DB cs.AIcited by 9

Not on the map. Why?

Abstract (arXiv)

In this paper, we present our vision of so called zero-shot learning for databases which is a new learning approach for database components. Zero-shot learning for databases is inspired by recent advances in transfer learning of models such as GPT-3 and can support a new database out-of-the box without the need to train a new model. Furthermore, it can easily be extended to few-shot learning by further retraining the model on the unseen database. As a first concrete contribution in this paper, we show the feasibility of zero-shot learning for the task of physical cost estimation and present very promising initial results. Moreover, as a second contribution we discuss the core challenges related to zero-shot learning for databases and present a roadmap to extend zero-shot learning towards many other tasks beyond cost estimation or even beyond classical database systems and workloads.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 21 of its 34 references that are in the map. First posted on arXiv 3 May 2021.

Builds on

21 of 34 references.

References · 21

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2017Deep Setscited by 1,421
  3. 2017The Case for Learned Index Structurescited by 188
  4. 2016Value Iteration Networkscited by 160
  5. 2018Learning Scheduling Algorithms for Data Processing Clusterscited by 134
  6. 2019Neo: A Learned Query Optimizercited by 80
  7. 2018Learned Cardinalities: Estimating Correlated Joins with Deep Learningcited by 64
  8. 2019ALEX: An Updatable Adaptive Learned Indexcited by 43
  9. 2019An End-to-End Learning-based Cost Estimatorcited by 43
  10. 2019Deep Unsupervised Cardinality Estimationcited by 36
  11. 2018Learning to Optimize Join Queries With Deep Reinforcement Learningcited by 32
  12. 2018FITing-Tree: A Data-aware Index Structurecited by 30
9 more
  1. 2018Deep Reinforcement Learning for Join Order Enumerationcited by 25
  2. 2020Tsunami: A Learned Multi-dimensional Index for Correlated Data and Skewed Workloadscited by 25
  3. 2019Plan-Structured Deep Neural Network Models for Query Performance Predictioncited by 24
  4. 2020NeuroCard: One Cardinality Estimator for All Tablescited by 21
  5. 2019DeepDB: Learn from Data, not from Queries!cited by 20
  6. 2019Opportunistic View Materialization with Deep Reinforcement Learningcited by 6
  7. 2018Model-Free Control for Distributed Stream Data Processing using Deep Reinforcement Learningcited by 4
  8. 2021High-Dimensional Bayesian Optimization with Multi-Task Learning for RocksDBcited by 4
  9. 2020Towards a General Framework for ML-based Self-tuning Databasescited by 2

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 1; 2024: 4; 2025: 1.

9 papers.

  1. 2022FactorJoin: A New Cardinality Estimation Framework for Join Queriescited by 13
  2. 2024E2ETune: End-to-End Knob Tuning via Fine-tuned Generative Language Modelcited by 9
  3. 2024PRICE: A Pretrained Model for Cross-Database Cardinality Estimationcited by 8
  4. 2022Detect, Distill and Update: Learned DB Systems Facing Out of Distribution Datacited by 6
  5. 2023Tailoring Machine Learning for Process Miningcited by 4
  6. 2024CardBench: A Benchmark for Learned Cardinality Estimation in Relational Databasescited by 4
  7. 2025In-Context Adaptation to Concept Drift for Learned Database Operationscited by 3
  8. 2024Robust Plan Evaluation based on Approximate Probabilistic Machine Learningcited by 1
  9. 2022Learning-Based Data Storage [Vision] (Technical Report)cited by 0