AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

Federated Dataset Simulation: Inducing Label-Free Heterogeneity Across Tasks

Vasilis Siomos, Lam Ngo, Jonathan Passerat-Palmbach, Giacomo Tarroni

NeurIPS 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated learning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a method for simulating federated datasets with label-free heterogeneity across tasks, as an alternative to Dirichlet label-skew partitioning of centralized datasets.

Problem

Federated learning evaluation relies on small centralized classification datasets partitioned via Dirichlet label skew, which is a one-dimensional proxy for the multi-dimensional heterogeneity of real federated data.

Approach

Simulates federated datasets that induce heterogeneity beyond label distribution (label-free) across tasks, rather than partitioning labels with Dirichlet skew. Details beyond the abstract are unknown.

Experimental setup

Baselines: Dirichlet label-skew partitioning

Builds on: Dirichlet label-skew partitioning

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy