# Optimization Trade-offs in Asynchronous Federated Learning: A Stochastic Networks Approach Source: [Optimization Trade-offs in Asynchronous Federated Learning: A Stochastic Networks Approach]() Collection index: [llms.txt]() Data updated: 2026-03-31T05:00:04.048Z Authors: Abdelkrim Alahyane, C\\'eline Comte, Matthieu Jonckheere Published: 2026-03-27T09:53:53.000Z ## Abstract Synchronous federated learning scales poorly due to the straggler effect. Asynchronous algorithms increase the update throughput by processing updates upon arrival, but they introduce two fundamental challenges: gradient staleness, which degrades convergence, and bias toward faster clients under heterogeneous data distributions. Although algorithms such as AsyncSGD and Generalized AsyncSGD mitigate this bias via client-side task queues, most existing analyses neglect the underlying queueing dynamics and lack closed-form characterizations of the update throughput and gradient staleness. To close this gap, we develop a stochastic queueing-network framework for Generalized AsyncSGD that jointly models random computation times at the clients and the central server, as well as random uplink and downlink communication delays. Leveraging product-form network theory, we derive a closed-form expression for the update throughput, alongside closed-form upper bounds for both the communication round complexity and the expected wall-clock time required to reach an $\\epsilon$-stationary point. These results formally characterize the trade-off between gradient staleness and wall-clock convergence speed. We further extend the framework to quantify energy consumption under stochastic timing, revealing an additional trade-off between convergence speed and energy efficiency. Building on these analytical results, we propose gradient-based optimization strategies to jointly optimize routing and concurrency. Experiments on EMNIST demonstrate reductions of 29%--46% in convergence time and 36%--49% in energy consumption compared to AsyncSGD. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on EMNIST IID (test) Efficiency Trade-offs](): Image Classification; EMNIST iid (test); Time Reduction (%) @ Acc 0.6: 67.5. - [Image Classification on EMNIST Non-IID (test)](): Image Classification; EMNIST Non-IID (test); Time Reduction (%) @ Acc 0.6: 79.3. - [Image Classification on KMNIST (test)](): Image Classification; KMNIST (test); Time Reduction: 18.95. - [Image Classification on CIFAR-100 (test) (Time Reduction Metrics)](): Image Classification; CIFAR-100 (test); Time Reduction (%) vs Round-Optimized: 67.95. ## Original sources and implementation [Original paper]() [PDF]()