AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

SPACE: Single-round Participant Amalgamation for Contribution Evaluation in Federated Learning

Yi-Chung Chen, Hsi-Wen Chen, Shun-Guei Wang, Ming-Syan Chen

NeurIPS 2023 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models · 5 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
single-round amalgamation instead of GTG-ShapleySPACE · MNISTcommunication cost (MB): 35.2 -> 1.76Improved
SPACE instead of TMC-ShapleySPACE · MNISTexecution time (s): 84796 -> 160Improved
SPACE instead of DIG-FLSPACE · MNIST Non-IIDPearson correlation: 0.8715 -> 0.9448Improved
SPACE instead of GTG-ShapleySPACE · CIFAR10 MislabelPearson correlation: 0.4148 -> 0.9641Improved
knowledge amalgamation instead of FedAvg aggregation (ablation)SPACE · MNIST Non-IIDPearson correlation: 0.9713 -> 0.9448Hurt
knowledge amalgamation instead of FedAvg aggregation (ablation)SPACE · MNISTexecution time (s): 294 -> 160Improved
knowledge amalgamation instead of FedAvg aggregation (ablation)SPACE · Tiny-ImageNetexecution time (s): 252 -> 453Hurt
hierarchical amalgamation with G=2 (ablation)SPACE · MNISTruntime (s): 160 -> 633Hurt
hierarchical amalgamation with G=2 (ablation)SPACE · MNIST MislabelPearson correlation: 0.9612 -> 0.9726No clear effect
hierarchical amalgamation with G=2 (ablation)SPACE · CIFAR10 Non-IIDPearson correlation: 0.9290 -> 0.9333No clear effect
replace labeled validation set with unlabeled data (ablation)SPACE · MNIST Non-IIDPearson correlation: 0.9448 -> 0.8772Hurt
replace labeled validation set with unlabeled data (ablation)SPACE · CIFAR10 MislabelPearson correlation: 0.9641 -> 0.9641No clear effect

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