Federated Learning with Profile Mapping under Distribution Shifts and Drifts

2026-02-07

AI Summary

This work addresses the significant performance degradation of federated learning under client data distribution shifts and dynamic drifts, a challenge exacerbated by existing methods that often rely on unrealistic prior assumptions—such as known cluster counts—and consequently exhibit limited generalization. To overcome these limitations, we propose Feroma, a novel framework that, for the first time, jointly handles both distribution shift and drift without requiring client identities or clustering priors. Feroma leverages privacy-preserving distributional profiling to enable similarity-based adaptive weighting, dynamically selects aggregation strategies, and supports zero-shot model assignment at test time. Evaluated across six benchmarks, Feroma achieves up to a 12-percentage-point improvement in average accuracy over ten state-of-the-art methods, while maintaining computational and communication overhead comparable to FedAvg, thereby substantially enhancing performance and stability in dynamic heterogeneous environments.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Distributed Search
• Multiagent Systems: Adversarial Agents

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
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated Learning (FL) enables decentralized model training across clients without sharing raw data, but its performance degrades under real-world data heterogeneity. Existing methods often fail to address distribution shift across clients and distribution drift over time, or they rely on unrealistic assumptions such as known number of client clusters and data heterogeneity types, which limits their generalizability. We introduce Feroma, a novel FL framework that explicitly handles both distribution shift and drift without relying on client or cluster identity. Feroma builds on client distribution profiles-compact, privacy-preserving representations of local data-that guide model aggregation and test-time model assignment through adaptive similarity-based weighting. This design allows Feroma to dynamically select aggregation strategies during training, ranging from clustered to personalized, and deploy suitable models to unseen, and unlabeled test clients without retraining, online adaptation, or prior knowledge on clients'data. Extensive experiments show that compared to 10 state-of-the-art methods, Feroma improves performance and stability under dynamic data heterogeneity conditions-an average accuracy gain of up to 12 percentage points over the best baselines across 6 benchmarks-while maintaining computational and communication overhead comparable to FedAvg. These results highlight that distribution-profile-based aggregation offers a practical path toward robust FL under both data distribution shifts and drifts.

Problem

• Federated Learning
• Distribution Shift
• Distribution Drift
• Data Heterogeneity
• Client Heterogeneity

Innovation

• Federated Learning
• Distribution Shift
• Distribution Drift
• Profile Mapping
• Adaptive Aggregation