AI Paper Atlas

Research map / Federated learning and privacy / Communication-efficient FL

FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training

Yijiang Li, Emon Dey, Zilinghan Li, Krishnan Raghavan, Ravi K Madduri, Kibaek Kim

ICML 2026 · Federated learning and privacy › Communication-efficient FL · approach: Client participation analysis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add EWMA queue-delay prediction (ablation)FedQueue · Synthetic non-IID MNISTtime to 85% accuracy: 116.59 -> 26.68Improved
add EWMA queue-delay prediction (ablation)FedQueue · Synthetic non-IID MNISTfinal accuracy: 96.01 -> 96.62No clear effect
add staleness decay in aggregation (ablation)FedQueue · Synthetic non-IID MNISTfinal accuracy: 87.18 -> 96.62Improved
add inverse learning-rate scaling (ablation)FedQueue · Synthetic non-IID MNISTfinal accuracy: 90.89 -> 96.62Improved
increase safety buffer delta (ablation)FedQueue · Synthetic queuestime to 95% accuracy and stalenessMixed
use EWMA rate alpha 0.5 instead of 0.1 or 1.0 (ablation)FedQueue · Synthetic queueslate arrivalsImproved
use queue-aware FedQueue instead of FedCompass/FedAvgfederated learning · Real HPC deployment, LLaMA2-7Bfinal test loss: 0.4345 -> 0.3947Improved
use FedQueue instead of baselines under high queue variancefederated learning · Synthetic non-IID MNIST, rho=0.9time to 95% accuracy: 245.60 -> 154.72 (vs FedAvg)Improved
use FedQueue instead of FedBuff/FedCompassfederated learning · Synthetic non-IID MNIST, rho=0.9max accuracy: 97.45 -> 96.62 (vs FedBuff)Hurt
use FedQueue instead of FedBuff/FedCompassfederated learning · CIFAR-100 IIDmax accuracy: 56.4 -> 51.7 (vs FedBuff)Hurt
use FedQueue instead of baselinesfederated learning · CIFAR-100 non-IIDmax accuracy: 44.7 -> 46.9 (vs FedCompass)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy