Newest papers

Learning to Refer: Inference-Time Collaboration in Decentralized Federated Learning

Yuheng Ma, Chendi Wang, Qiong Zhang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26800 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

Na aprendizagem federada descentralizada de dados, os clientes heterogêneos podem desenvolver competências complementares, mas cada consulta é tipicamente avaliada apenas pelo cliente onde chega. Para explorar essa competência distribuída sem coordenação global, formulamos a colaboração de inferência-tempo como referência especializada local orçamentada, um processo de roteamento limitado por gráficos no qual uma consulta só pode se mover entre clientes vizinhos e apenas clientes visitados contribuem para sua previsão final. O EdgeDRM implementa este quadro aprendendo um potencial de roteamento baseado em densidade para cada cliente: as diferenças entre potenciais vizinhos fornecem as comparações usadas para selecionar referências sucessivas, enquanto os modelos de tarefas locais geram previsões que são agregadas ao longo da rota resultante. Tanto o treinamento quanto a inferência são descentraliz

Machine translation (Meta M2M100); the English original follows.

Abstract

In decentralized federated learning, heterogeneous clients may develop complementary expertise, yet each query is typically evaluated only by the client where it arrives. To exploit this distributed expertise without global coordination, we formulate inference-time collaboration as budgeted local expert referral, a graph-constrained routing process in which a query can move only between neighboring clients and only visited clients contribute to its final prediction. EdgeDRM implements this framework by learning a density-based routing potential for each client: differences between neighboring potentials provide the comparisons used to select successive referrals, while local task models generate predictions that are aggregated along the resulting route. Both training and inference are decentralized, using only neighbor-local information without a central router or graph-wide client scoring. We establish estimation guarantees for the learned comparisons and characterize how local improvement and graph structure govern prior-weighted density gains from referral. Experiments on benchmark and real-world medical data show that EdgeDRM outperforms graph-matched decentralized baselines in most settings, with the clearest benefits under sparse connectivity.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Exploit, Inference, Task (project management), Process (computing), Benchmark (surveying)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange