From Models to Systems: A Comprehensive Fairness Framework for Compositional Recommender Systems
This addresses fairness challenges in real-world recommendation systems for developers and regulators, though it is incremental by adapting existing optimization tools.
The paper tackles the problem of ensuring fairness in compositional recommender systems by moving beyond individual model audits to a holistic system-level framework, and demonstrates its effectiveness through empirical results on synthetic and real datasets.
Fairness research in machine learning often centers on ensuring equitable performance of individual models. However, real-world recommendation systems are built on multiple models and even multiple stages, from candidate retrieval to scoring and serving, which raises challenges for responsible development and deployment. This system-level view, as highlighted by regulations like the EU AI Act, necessitates moving beyond auditing individual models as independent entities. We propose a holistic framework for modeling system-level fairness, focusing on the end-utility delivered to diverse user groups, and consider interactions between components such as retrieval and scoring models. We provide formal insights on the limitations of focusing solely on model-level fairness and highlight the need for alternative tools that account for heterogeneity in user preferences. To mitigate system-level disparities, we adapt closed-box optimization tools (e.g., BayesOpt) to jointly optimize utility and equity. We empirically demonstrate the effectiveness of our proposed framework on synthetic and real datasets, underscoring the need for a system-level framework.