Skip to content
Book Open access

Taming Update Drift in Asynchronous Federated Learning via Orthogonal Calibration

Aug 2026 · Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 · pp. 6464-6475 · 0 citations · 59 references

TL;DR

OrthoFL is proposed, an orthogonal calibration framework that decouples global and local learning progress to reduce interference and consistently outperforms state-of-the-art asynchronous baselines under various delay patterns and heterogeneity scenarios.

Abstract

Asynchronous federated learning improves the efficiency of conventional synchronous protocols by integrating updates as they arrive. However, asynchrony and data heterogeneity make learning objectives at global and local levels inherently inconsistent—global optimization trajectories can conflict with ongoing local updates. Existing methods simply distribute the latest global weights to clients, which may overwrite local progress and cause model drift. We propose OrthoFL, an orthogonal calibration framework that decouples global and local learning progress to reduce interference. Upon receiving an update, the server aggregates it into the global weights via an adaptive moving average. For local training, OrthoFL computes the global shift due to updates from other clients during the client's delay and removes its projection onto the received update. The resulting parameters lie in a subspace orthogonal to the client update and preserve the closest component of global progress within the orthogonal hyperplane. The calibrated shift is then merged into the client model for further training. Extensive experiments across five datasets demonstrate that OrthoFL improves accuracy by 7.4 percentage points and achieves up to a 12× speedup over the best-performing synchronous baseline. Moreover, it consistently outperforms state-of-the-art asynchronous baselines under various delay patterns and heterogeneity scenarios.

Read PDF

Similar papers

#machine learning Preprint Sep 2026

Robust Decentralized Personalized Federated Learning via Prediction-Constrained Neighborhood Collaboration

This paper proposes a robust decentralized personalized federated learning method R-DPFL, that enables clients to reduce the impact of Byzantine attacks via robust neighborhood direction estimation and history-based update trend prediction, rather than purely aggregating client models as in the existing work. In R-DPFL...

Xiao Ma, Hong Shen, Hui Tian et al. · 1 citation
Open access Oct 2026

Taming Staleness in Asynchronous Federated Learning With System Heterogeneity

This work proposes FedASRA, an efficient semi-asynchronous federated learning framework featuring adaptive alternating selection (AAS) and real-time staleness adjustment (RSA), and provides convergence and system complexity analysis for FedASRA, establishing a staleness-agnostic convergence upper bound.

Zhen-Hao Liu, Meng-Yao Li, Bo Zhang et al. · 1 citation

FedRAW: Preserving Rare-Label Influence in Asynchronous Federated Learning

FedRAW is proposed, a fully server-side aggregation method that preserves rare-label influence without changing local training, client objectives, or communication protocols, and improves rarelabel accuracy while preserving comparable global accuracy and adding negligible server-side computation.

Prashant Bajpai, Divya Saxena, Philippe Lalanda et al. · 0 citations
#machine learning Preprint Sep 2026

Resilience Beyond Stationary Client Unavailability: Unlocking Efficient and Unbiased Federated Learning

FedSWE is proposed, which admits novel algorithmic structures to compensate for missed computations, stabilize and diffuse the global updates over rounds, and evenly mix the local updates through implicit gossiping, despite being agnostic to non-stationary dynamics.

Ming Xiang, Stratis Ioannidis, Edmund Yeh et al. · 0 citations
#machine learning Preprint Oct 2026

A Path Integral Surrogate for Multi-Step Gradient Inversion in Federated Learning

Federated learning lets many clients train a shared model together without ever sending their private data to a central server. Each client shares only a model update, and this update should reveal far less about the client than its raw training examples would. This premise is what protects the privacy of the clients....

Agnivo Ghosh, Saumik Bhattacharya · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.