Summary
Federated learning lets multiple users train an AI together without sharing their private data, but it can be hard to know if everyone is following the rules or being fair. The authors propose a new system called Governance-Aware Federated Learning (GAFL) that checks evidence of good behavior, fairness, and trust for each participant during training. They tested GAFL with 10 participants and found it was able to detect bad or dishonest actions effectively, though this added some processing time and made fairness harder to balance. This work shows how AI collaboration can be monitored better to build more trustworthy shared models.
What this means in practice
- •For enterprise ai teams: Monitor and enforce trust and fairness policies in collaborative AI model training across multiple company branches or departments.
- •For cloud service operators: Provide federated learning platforms with built-in governance that detects and intervenes on malicious or non-compliant clients during training.$Commercial implications: Enables cloud providers to sell federated learning services with enhanced security and compliance features.
Tested on one dataset.
Authors
Kiranmayee Janardhan, Karthick S, Madhusmita Panda, Bharathi Vokkaliga Ganesh, Syed Ahmed
Abstract
Federated Learning (FL) enables collaborative model training without exchanging raw data, but conventional FL provides limited runtime governance, evidence verification, fairness assessment, explainability monitoring, and auditability. This paper proposes Governance-Aware Federated Learning (GAFL), which combines independent governance-evidence verification, temporal trust, fairness and explanation assessment, anomaly detection, policy enforcement, normalized aggregation, and tamper-evident audit logging. Client-reported governance values are authenticated, bound to their submitted model updates, and independently recomputed before aggregation. A cross-silo prototype with 10 participants was evaluated on the Adult Income dataset under honest, benign non-IID, biased-data, malicious-update, non-compliant, falsified-evidence, and mixed-threat scenarios. Four methods were compared across five random seeds: FedAvg, FedAvg with anomaly filtering, GAFL without independent verification, and complete GAFL. Complete GAFL intervened on all affected participants at the final round across the evaluated risk scenarios, including identifying 100% of falsified-evidence participants, whereas GAFL without verification identified none of the falsified-evidence participants. Final macro-F1 ranged from 0.7520 to 0.7733 across GAFL scenarios. Complete GAFL required 173.67 plus or minus 26.94 ms of governance processing per round, compared with 3.82 plus or minus 0.79 ms for FedAvg. A strict demographic-parity threshold also caused frequent interventions for honest participants, demonstrating a significant fairness-policy and utility trade-off.