Research in Collaborative Learning Does Not Serve Cross-Silo Federated Learning in Practice

2025-10-14

AI Summary

Cross-institutional federated learning (FL) faces significant deployment barriers in practice, as existing research predominantly focuses on technical optimization while neglecting socio-technical challenges inherent in inter-organizational collaboration. Through in-depth interviews with user organizations, software vendors, and academic researchers, this study identifies three core challenges: model performance imbalance across institutions, lack of cross-institutional incentives, and absence of robust trust mechanisms. We find that cross-institutional FL fundamentally differs from cross-device FL—its success hinges more on institutional design than algorithmic refinement alone. Current collaborative learning research fails to address critical real-world governance, accountability, and sustainability requirements. This work is the first to systematically analyze cross-institutional FL through a socio-technical lens, revealing its practical bottlenecks. It advocates a paradigm shift in FL research toward trust-building, incentive-compatible mechanisms, and organizational alignment—providing both theoretical foundations and actionable pathways for transitioning FL from lab-scale experiments to industry-wide collaborative deployment.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Multiagent Systems: Other Foundations of Multi Agent Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Social Networks and Social Media: Socio-technical web-based systems

Abstract

Cross-silo federated learning (FL) is a promising approach to enable cross-organization collaboration in machine learning model development without directly sharing private data. Despite growing organizational interest driven by data protection regulations such as GDPR and HIPAA, the adoption of cross-silo FL remains limited in practice. In this paper, we conduct an interview study to understand the practical challenges associated with cross-silo FL adoption. With interviews spanning a diverse set of stakeholders such as user organizations, software providers, and academic researchers, we uncover various barriers, from concerns about model performance to questions of incentives and trust between participating organizations. Our study shows that cross-silo FL faces a set of challenges that have yet to be well-captured by existing research in the area and are quite distinct from other forms of federated learning such as cross-device FL. We end with a discussion on future research directions that can help overcome these challenges.

Problem

• Identifying practical barriers to cross-silo federated learning adoption
• Analyzing stakeholder concerns about model performance and incentives
• Distinguishing challenges from other federated learning approaches

Innovation

• Interview study identifies cross-silo FL adoption barriers
• Analysis reveals model performance and trust concerns
• Distinct challenges compared to cross-device FL highlighted