# Personalized Federated Learning on FEMNIST (test) Source: [Personalized Federated Learning on FEMNIST (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:06:35.238Z Evaluation of Personalized Federated Learning performance on the test split of the FEMNIST dataset, focusing on metrics that capture performance variance across clients. 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. | Scope | Recorded value | | --- | --- | | Task | Personalized Federated Learning | | Dataset (including any stated split) | FEMNIST (test) | | Metric | Top 10% Most Samples Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Top 10% Most Samples Accuracy](): Higher is better; unit: PERCENT. - [Worst 10% User Avg Acc](): Higher is better; unit: PERCENT. ## Recorded results | Method | Top 10% Most Samples Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Self-FL | 76.24 | Not recorded | 68fa6c5c1e209131cea3395a | | PerFedAvg | 74.95 | Not recorded | 68fa6c5c1e209131cea3395a | | pFedMe | 72.76 | Not recorded | 68fa6c5c1e209131cea3395a | | DITTO | 70.58 | Not recorded | 68fa6c5c1e209131cea3395a | | FedAvg | 70.48 | Not recorded | 68fa6c5c1e209131cea3395a | ### Self-FL Top 10% Most Samples Accuracy: 76.24. Result ID: 698f0b953a6c314849f536b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"local training steps":"adaptive","maximum local steps (l\_max)":"40","batch size":"10","learning rate":"0.01"}. This record does not assert independent reproduction. ### PerFedAvg Top 10% Most Samples Accuracy: 74.95. Result ID: 698f0b953a6c314849f536b7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"local training steps":"20","batch size":"10","learning rate":"0.01"}. This record does not assert independent reproduction. ### pFedMe Top 10% Most Samples Accuracy: 72.76. Result ID: 698f0b953a6c314849f536b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"local training steps":"20","batch size":"10","learning rate":"0.01"}. This record does not assert independent reproduction. ### DITTO Top 10% Most Samples Accuracy: 70.58. Result ID: 698f0b953a6c314849f536b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"local training steps":"20","batch size":"10","learning rate":"0.01"}. This record does not assert independent reproduction. ### FedAvg Top 10% Most Samples Accuracy: 70.48. Result ID: 698f0b953a6c314849f536b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"local training steps":"20","batch size":"10","learning rate":"0.01"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.