mlsanityData to 14 Oct 2026

QueryMarket: Cost-Aware Online Active Learning in Data Markets

arXiv 2606.17805 ↗First posted 16 Jun 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Data acquisition is a major bottleneck for learning in real-time streams: analysts must decide on the fly which labels to purchase while respecting a rolling budget. However, existing online active learning rarely unifies pricing, information gain, and rolling budget constraints under concept drift. We introduce QueryMarket, a market-inspired framework that queries each incoming data point based on its estimated utility to the model and its price. Within this framework, we propose OVBAL (online variance-based active learning), which integrates data pricing with information-driven selection by estimating each sample's marginal utility via a D-optimality criterion with exponential forgetting and executing cost-aware purchases under rolling budget constraints. OVBAL yields a simple, fully online decision rule that adapts to nonstationary streams and heterogeneous label costs. Experiments on synthetic data and a real-world solar power generation forecasting task show that OVBAL is particularly effective under seller-centric pricing and yields a more favorable long-run error-cost trade-off in the real-world task under both pricing schemes.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 19 of its 35 references that are in the map. First posted on arXiv 16 Jun 2026.

Builds on

19 of 35 references.

References · 19

  1. 2019Data Shapley: Equitable Valuation of Data for Machine Learningcited by 629
  2. 2019Online Learning: A Modern Introduction Using Convex Optimizationcited by 417
  3. 2018Online Learning: A Comprehensive Surveycited by 205
  4. 2019Data Valuation using Reinforcement Learningcited by 116
  5. 2024A Survey on Deep Active Learning: Recent Advances and New Frontierscited by 54
  6. 2016Efficient Second Order Online Learning by Sketchingcited by 33
  7. 2023Active learning for data streams: a surveycited by 29
  8. 2023Navigating the Pitfalls of Active Learning Evaluation: A Systematic Framework for Meaningful Performance Assessmentcited by 26
  9. 2023Streaming Active Learning with Deep Neural Networkscited by 20
  10. 2022Fundamentals of Task-Agnostic Data Valuationcited by 19
  11. 2021Regression markets and application to energy forecastingcited by 11
  12. 2022Stream-based active learning with linear modelscited by 9
7 more
  1. 2014Active Regression by Stratificationcited by 8
  2. 2022Online Active Learning for Soft Sensor Development using Semi-Supervised Autoencoderscited by 6
  3. 2016Online Active Linear Regression via Thresholdingcited by 5
  4. 2022Strategic Coalition for Data Pricing in IoT Data Marketscited by 3
  5. 2024Composite Active Learning: Towards Multi-Domain Active Learning with Theoretical Guaranteescited by 3
  6. 2025Budgeted Online Active Learning with Expert Advice and Episodic Priorscited by 1
  7. 2025How to Purchase Labels? A Cost-Effective Approach Using Active Learning Marketscited by 1

Built on by

Not cited yet.