Metraipolis › archive › dc_643b15b11bb9
Understanding Trajectory Heterogeneity in Federated World Model Learning — arXiv:2610.02957
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.02957
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T22:51:57.982Z
- SHA-256 of the archived text
04735dd100b3cabe05c391ddc2ce34fe04b5ff3d2343a7fb58ca8d764b71c488- Stable URL of this page
https://metraipolis.com/documento/dc_643b15b11bb9
How to cite
Understanding Trajectory Heterogeneity in Federated World Model Learning — arXiv:2610.02957 — arXiv, https://arxiv.org/abs/2610.02957 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_643b15b11bb9 (Metraipolis archive dc_643b15b11bb9, sha256 04735dd100b3cabe05c391ddc2ce34fe04b5ff3d2343a7fb58ca8d764b71c488)
BibTeX
@misc{dc643b15b11bb9,
title = {Understanding Trajectory Heterogeneity in Federated World Model Learning — arXiv:2610.02957},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.02957},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_643b15b11bb9; sha256 04735dd100b3cabe05c391ddc2ce34fe04b5ff3d2343a7fb58ca8d764b71c488},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02957. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
World models learn state evolution from trajectories, making access to temporal context a central training requirement. Federated learning can use distributed records, while ownership boundaries within a trajectory restrict the examples each client can construct. Our study benchmarks this cross-time setting through hourly action-conditioned clinical prediction on eight MIMIC-IV disease cohorts, comprising 40.87 million transition memberships. We specify severity-based client ownership, patient-separated construction, local history and future-window rules, and paired rollout evaluation from one to 32 hours. A matrix of ten federated algorithms covers 32 disease--partition configurations under five rounds of ten-percent participation. Three findings emerge from existing results and training logs. First, client ownership and participation jointly restrict long-window coverage: only 7.55\%--21.36\% of pooled-available 32-step windows have a locally complete anchor visited during training, averaged across diseases. Second, finer severity partitions accompany higher FedAvg error in 15 of 16 paired comparisons, while algorithm gains are small and horizon-dependent: FedProx reduces mean error by 0.56\%, with no consistent improvement at 32 steps. Third, algorithm labels conceal distinct update behavior, including inactive extrapolation and orders-of-magnitude differences in update scale. Cached-update performance also varies strongly across trajectory partitions under the same benchmark protocol. These results establish temporal access, participation coverage, optimization behavior, and horizon-resolved prediction as complementary dimensions for evaluating federated clinical world models. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.02957 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-05T22:46:10.618Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).