Heterogeneous Data Game: Characterizing the Model Competition Across Multiple Data Sources
This addresses the challenge of competition in ML marketplaces with heterogeneous data, offering insights for regulatory policies and practical strategies, though it is incremental as it builds on existing game-theoretic and data heterogeneity concepts.
The paper tackles the problem of multiple competing ML providers in markets with heterogeneous data sources by proposing a game-theoretic framework, the Heterogeneous Data Game, and shows that pure Nash equilibria can be non-existent, homogeneous, or heterogeneous, with providers converging on the same model or specializing in distinct data sources.
Data heterogeneity across multiple sources is common in real-world machine learning (ML) settings. Although many methods focus on enabling a single model to handle diverse data, real-world markets often comprise multiple competing ML providers. In this paper, we propose a game-theoretic framework -- the Heterogeneous Data Game -- to analyze how such providers compete across heterogeneous data sources. We investigate the resulting pure Nash equilibria (PNE), showing that they can be non-existent, homogeneous (all providers converge on the same model), or heterogeneous (providers specialize in distinct data sources). Our analysis spans monopolistic, duopolistic, and more general markets, illustrating how factors such as the "temperature" of data-source choice models and the dominance of certain data sources shape equilibrium outcomes. We offer theoretical insights into both homogeneous and heterogeneous PNEs, guiding regulatory policies and practical strategies for competitive ML marketplaces.