analiza aiqbench
This research addresses a critical bottleneck in edge AI by optimizing data admission strategies for streaming federated learning under strict memory constraints. The authors propose an Active-Constraint Drift-Plus-Penalty policy that mathematically balances sampling costs against model accuracy, offering sublinear regret guarantees essential for scalable distributed training systems. By characterizing effective sample size and reuse imbalance, this work provides a rigorous framework for managing limited client resources without sacrificing learning performance.
Adaptive Data Admission and Retention for Streaming Federated Learning
arXiv:2607.23987v1 Announce Type: cross Abstract: We study streaming federated learning with limited client memory, where newly generated training data incur time-varying sampling costs and must be selectively admitted and retained over time. We consider a joint server-side admission and client-side memory-management framework with the objective of minimizing the cumulative excess population risk under a sampling-cost budget and buffer constraints. We first derive a learning-error bound that explicitly captures the effects of instantaneous training sample size, distinct-sample growth, and reuse imbalance through a characterization of the effective sample size. Through a surrogate penalty obtained from this bound, we develop an Active-Constraint Drift-Plus-Penalty (ACDPP) policy that combines a structured client-side $K$-step retention rule with a server-side online admission rule and a time-varying rectangular admission region. We further present a sequence of comparison arguments, via an auxiliary constant-admission policy, that connects the ACDPP learning bound to a costless oracle benchmark. This yields explicit guarantees in terms of sublinear regret and sampling-cost violation, while the buffer-occupancy violation is controlled through offline selection of the retention horizon. Experiments on multiple datasets demonstrate that the proposed policy remains close to the oracle benchmark while satisfying the sampling-cost and buffer constraints.
Abstract:We study streaming federated learning with limited client memory, where newly generated training data incur time-varying sampling costs and must be selectively admitted and retained over time. We consider a joint server-side admission and client-side memory-management framework with the objective of minimizing the cumulative excess population risk under a sampling-cost budget and buffer constraints. We first derive a learning-error bound that explicitly captures the effects of instantaneous training sample size, distinct-sample growth, and reuse imbalance through a characterization of the effective sample size. Through a surrogate penalty obtained from this bound, we develop an Active-Constraint Drift-Plus-Penalty (ACDPP) policy that combines a structured client-side $K$-step retention rule with a server-side online admission rule and a time-varying rectangular admission region. We further present a sequence of comparison arguments, via an auxiliary constant-admission policy, that connects the ACDPP learning bound to a costless oracle benchmark. This yields explicit guarantees in terms of sublinear regret and sampling-cost violation, while the buffer-occupancy violation is controlled through offline selection of the retention horizon. Experiments on multiple datasets demonstrate that the proposed policy remains close to the oracle benchmark while satisfying the sampling-cost and buffer constraints.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI)
arXiv:2607.23987 [cs.LG]
(or arXiv:2607.23987v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.23987 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 291 cuvinte
Subiecte conexe
Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning
arXiv · 7 oct. 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · 27 aug. 2026
Building Federated Multimodal AI Workflows with NVIDIA FLARE
NVIDIA · 19 aug. 2026
Federated continual learning: A comprehensive survey on lifelong and privacy-preserving learning over distributed and non-stationary data
arXiv · 12 iun. 2026