Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

One-Shot Federated Learning

About

We present one-shot federated learning, where a central server learns a global model over a network of federated devices in a single round of communication. Our approach - drawing on ensemble learning and knowledge aggregation - achieves an average relative gain of 51.5% in AUC over local baselines and comes within 90.1% of the (unattainable) global ideal. We discuss these methods and identify several promising directions of future work.

Neel Guha, Ameet Talwalkar, Virginia Smith• 2019

Related benchmarks

TaskDatasetResultRank
Image ClassificationCIFAR-100 (test)
Accuracy32.25
3518
Image ClassificationCIFAR-10 (test)
Accuracy79.91
3381
Image ClassificationSVHN (test)
Accuracy85.7
199
ClassificationfMNIST (test)
Accuracy66.19
149
Graph ClassificationENZYMES 1.0 (test)
AUC49
25
Graph ClassificationIMDB-BINARY 1.0 (test)
AUC0.54
25
Graph ClassificationIMDB-MULTI (IMDB-M) 1.0 (test)
AUC52
25
Graph ClassificationPROTEINS 1.0 (test)
AUC0.76
17
Graph ClassificationMUTAG 1.0 (test)
AUC0.72
17
Image ClassificationTiny-ImageNet
Accuracy (alpha=0.1)30.85
9
Showing 10 of 15 rows

Other info

Follow for update