{"ID":2870062,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.14444","arxiv_id":"2509.14444","title":"FedAVOT: Exact Distribution Alignment in Federated Learning via Masked Optimal Transport","abstract":"Federated Learning (FL) allows distributed model training without sharing raw data, but suffers when client participation is partial. In practice, the distribution of available users (\\emph{availability distribution} $q$) rarely aligns with the distribution defining the optimization objective (\\emph{importance distribution} $p$), leading to biased and unstable updates under classical FedAvg. We propose \\textbf{Fereated AVerage with Optimal Transport (\\textbf{FedAVOT})}, which formulates aggregation as a masked optimal transport problem aligning $q$ and $p$. Using Sinkhorn scaling, \\textbf{FedAVOT} computes transport-based aggregation weights with provable convergence guarantees. \\textbf{FedAVOT} achieves a standard $\\mathcal{O}(1/\\sqrt{T})$ rate under a nonsmooth convex FL setting, independent of the number of participating users per round. Our experiments confirm drastically improved performance compared to FedAvg across heterogeneous, fairness-sensitive, and low-availability regimes, even when only two clients participate per round.","short_abstract":"Federated Learning (FL) allows distributed model training without sharing raw data, but suffers when client participation is partial. In practice, the distribution of available users (\\emph{availability distribution} $q$) rarely aligns with the distribution defining the optimization objective (\\emph{importance distribu...","url_abs":"https://arxiv.org/abs/2509.14444","url_pdf":"https://arxiv.org/pdf/2509.14444v1","authors":"[\"Herlock\",\"Rahimi\",\"Dionysis Kalogerias\"]","published":"2025-09-17T21:35:16Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"stat.ML\"]","methods":"[]","has_code":false}