Metraipolis › archive › dc_d48b2cf06002

DART-FL: Burst-Aware Multitask Federated Learning under Dynamic Inference Demand at the Edge — arXiv:2608.27713

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.27713
License (as archived)
CC0-1.0
Archived
2026-10-17T11:06:24.710Z
SHA-256 of the archived text
bcf7dd520470b2a0eb1819cebb566bf39e1b3a0ce73639e11d4351617aaae5e1
Stable URL of this page
https://metraipolis.com/documento/dc_d48b2cf06002

How to cite

DART-FL: Burst-Aware Multitask Federated Learning under Dynamic Inference Demand at the Edge — arXiv:2608.27713 — arXiv (en), https://arxiv.org/abs/2608.27713 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d48b2cf06002 (Metraipolis archive dc_d48b2cf06002, sha256 bcf7dd520470b2a0eb1819cebb566bf39e1b3a0ce73639e11d4351617aaae5e1)
BibTeX
@misc{dcd48b2cf06002,
  title = {DART-FL: Burst-Aware Multitask Federated Learning under Dynamic Inference Demand at the Edge — arXiv:2608.27713},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27713},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d48b2cf06002; sha256 bcf7dd520470b2a0eb1819cebb566bf39e1b3a0ce73639e11d4351617aaae5e1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27713. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Edge intelligence systems increasingly require model training and online inference to coexist on resource-constrained devices, while inference demand can vary substantially across tasks over time. This creates two coupled challenges: sufficient computation must be reserved for inference to maintain service-level objectives (SLOs), while the remaining training capacity should adapt to task-specific demand so that frequently requested tasks can improve earlier during training. We propose an SLO-aware, demand-driven multitask federated learning framework (DART-FL) that jointly adapts the inference-training resource split and task-level training emphasis. At each scheduling interval, DART-FL uses the inference backlog and profiled service capacity to determine the minimum resource allocation required for inference. The remaining training capacity is then distributed across tasks using a queue-aware DPP-inspired scheduler, and the resulting task allocations are mapped to dynamic loss weights. This allows tasks experiencing higher inference demand to receive greater training emphasis in earlier communication rounds. Clients train a shared backbone with task-specific heads, and the complete multitask model is aggregated through FedAvg. We evaluate DART-FL using Stanford Cars and Oxford Flowers 102 under both synthetic and real Alibaba trace-derived workloads. Results show that DART-FL dynamically adapts the inference-training resource split to time-varying inference demand and shifts the learning progress of high-demand tasks toward their burst periods, improving model accuracy when those tasks are frequently requested while maintaining comparable long-term multitask performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.27713 — 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 2026-10-17T10:50:25.770Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).