From Models to Systems: A Comprehensive Framework for AI System Fairness in Compositional Recommender Systems

Brian Hsu, Cyrus DiCiccio, Natesh S. Pillai, Hongseok Namkoong
Proceedings of the Algorithmic Fairness Through the Lens of Metrics and Evaluation, PMLR 279:8-37, 2025.

Abstract

Fairness research in machine learning often centers on ensuring equitable performance ofindividual models. However, real-world recommendation systems are built on multiplemodels and even multiple stages, from candidate retrieval to scoring and serving, whichraises challenges for responsible development and deployment. This AI system-level view,as highlighted by regulations like the EU AI Act, necessitates moving beyond auditingindividual models as independent entities. We propose a holistic framework for modelingAI system-level fairness, focusing on the end-utility delivered to diverse user groups, andconsider interactions between components such as retrieval and scoring models. We provideformal insights on the limitations of focusing solely on model-level fairness and highlight theneed for alternative tools that account for heterogeneity in user preferences. To mitigatesystem-level disparities, we adapt closed-box optimization tools (e.g., BayesOpt) to jointlyoptimize utility and equity. We empirically demonstrate the effectiveness of our proposedframework on synthetic and real datasets, underscoring the need for a framework thatreflects the design of modern, industrial AI systems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v279-hsu25a, title = {From Models to Systems: A Comprehensive Framework for AI System Fairness in Compositional Recommender Systems}, author = {Hsu, Brian and DiCiccio, Cyrus and Pillai, Natesh S. and Namkoong, Hongseok}, booktitle = {Proceedings of the Algorithmic Fairness Through the Lens of Metrics and Evaluation}, pages = {8--37}, year = {2025}, editor = {Rateike, Miriam and Dieng, Awa and Watson-Daniels, Jamelle and Fioretto, Ferdinando and Farnadi, Golnoosh}, volume = {279}, series = {Proceedings of Machine Learning Research}, month = {14 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v279/main/assets/hsu25a/hsu25a.pdf}, url = {https://proceedings.mlr.press/v279/hsu25a.html}, abstract = {Fairness research in machine learning often centers on ensuring equitable performance ofindividual models. However, real-world recommendation systems are built on multiplemodels and even multiple stages, from candidate retrieval to scoring and serving, whichraises challenges for responsible development and deployment. This AI system-level view,as highlighted by regulations like the EU AI Act, necessitates moving beyond auditingindividual models as independent entities. We propose a holistic framework for modelingAI system-level fairness, focusing on the end-utility delivered to diverse user groups, andconsider interactions between components such as retrieval and scoring models. We provideformal insights on the limitations of focusing solely on model-level fairness and highlight theneed for alternative tools that account for heterogeneity in user preferences. To mitigatesystem-level disparities, we adapt closed-box optimization tools (e.g., BayesOpt) to jointlyoptimize utility and equity. We empirically demonstrate the effectiveness of our proposedframework on synthetic and real datasets, underscoring the need for a framework thatreflects the design of modern, industrial AI systems.} }
Endnote
%0 Conference Paper %T From Models to Systems: A Comprehensive Framework for AI System Fairness in Compositional Recommender Systems %A Brian Hsu %A Cyrus DiCiccio %A Natesh S. Pillai %A Hongseok Namkoong %B Proceedings of the Algorithmic Fairness Through the Lens of Metrics and Evaluation %C Proceedings of Machine Learning Research %D 2025 %E Miriam Rateike %E Awa Dieng %E Jamelle Watson-Daniels %E Ferdinando Fioretto %E Golnoosh Farnadi %F pmlr-v279-hsu25a %I PMLR %P 8--37 %U https://proceedings.mlr.press/v279/hsu25a.html %V 279 %X Fairness research in machine learning often centers on ensuring equitable performance ofindividual models. However, real-world recommendation systems are built on multiplemodels and even multiple stages, from candidate retrieval to scoring and serving, whichraises challenges for responsible development and deployment. This AI system-level view,as highlighted by regulations like the EU AI Act, necessitates moving beyond auditingindividual models as independent entities. We propose a holistic framework for modelingAI system-level fairness, focusing on the end-utility delivered to diverse user groups, andconsider interactions between components such as retrieval and scoring models. We provideformal insights on the limitations of focusing solely on model-level fairness and highlight theneed for alternative tools that account for heterogeneity in user preferences. To mitigatesystem-level disparities, we adapt closed-box optimization tools (e.g., BayesOpt) to jointlyoptimize utility and equity. We empirically demonstrate the effectiveness of our proposedframework on synthetic and real datasets, underscoring the need for a framework thatreflects the design of modern, industrial AI systems.
APA
Hsu, B., DiCiccio, C., Pillai, N.S. & Namkoong, H.. (2025). From Models to Systems: A Comprehensive Framework for AI System Fairness in Compositional Recommender Systems. Proceedings of the Algorithmic Fairness Through the Lens of Metrics and Evaluation, in Proceedings of Machine Learning Research 279:8-37 Available from https://proceedings.mlr.press/v279/hsu25a.html.

Related Material