QueryMarket: Cost-Aware Online Active Learning in Data Markets

2026-06-16

AI Summary

This work addresses the challenge of dynamic sample selection under rolling budget constraints in real-time data streams. The authors propose QueryMarket, a novel framework that, for the first time, integrates data pricing, information gain, and rolling budgets within an online active learning paradigm. Its core algorithm, OVBAL, combines D-optimality criteria, an exponential forgetting mechanism, and variance-driven marginal utility estimation to enable cost-aware, fully online query decisions in the presence of concept drift and heterogeneous labeling costs. Experimental results demonstrate that the proposed method significantly outperforms baseline approaches on both synthetic datasets and a real-world solar power forecasting task, particularly excelling in seller-priced scenarios by effectively balancing long-term prediction error against annotation cost.

Abstract

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.

Problem

• online active learning
• data markets
• rolling budget
• concept drift
• label acquisition

Innovation

• online active learning
• cost-aware selection
• rolling budget
• concept drift
• D-optimality