🤖 AI Summary
Data silos deprive organizations of sufficient training data, while other entities hold underutilized data assets. To address this, we propose a data-augmented automated machine learning (AutoML) marketplace integrated with cloud-based AutoML platforms (e.g., Vertex AI, SageMaker), enabling automatic performance enhancement of buyer models via external data integration. We introduce a novel dynamic pricing mechanism grounded in *instrumental value*—defined as the marginal improvement in model performance attributable to external data—thereby circumventing direct data valuation challenges, mitigating strategic behavior, and supporting multi-tier menu-based pricing. Our approach unifies automated data discovery, model augmentation, performance attribution analysis, and incentive-compatible mechanism design within a unified cloud AutoML pipeline and economic modeling framework. Experiments demonstrate fair, robust, and scalable performance-driven pricing, significantly improving data utilization and marketplace revenue while enabling sustainable commercialization of external data.
📝 Abstract
Organizations often lack sufficient data to effectively train machine learning (ML) models, while others possess valuable data that remains underutilized. Data markets promise to unlock substantial value by matching data suppliers with demand from ML consumers. However, market design involves addressing intricate challenges, including data pricing, fairness, robustness, and strategic behavior. In this paper, we propose a pragmatic data-augmented AutoML market that seamlessly integrates with existing cloud-based AutoML platforms such as Google's Vertex AI and Amazon's SageMaker. Unlike standard AutoML solutions, our design automatically augments buyer-submitted training data with valuable external datasets, pricing the resulting models based on their measurable performance improvements rather than computational costs as the status quo. Our key innovation is a pricing mechanism grounded in the instrumental value - the marginal model quality improvement - of externally sourced data. This approach bypasses direct dataset pricing complexities, mitigates strategic buyer behavior, and accommodates diverse buyer valuations through menu-based options. By integrating automated data and model discovery, our solution not only enhances ML outcomes but also establishes an economically sustainable framework for monetizing external data.
