# Image Classification on EMNIST (val)

Source: [Image Classification on EMNIST (val)](<https://www.sota2.com/research/sota/image-classification-on-emnist-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:02:29.625Z

Evaluation of Image Classification performance on the validation split of the EMNIST dataset.

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 | Image Classification |
| Dataset (including any stated split) | EMNIST (val) |
| Metric | Rounds to 20.0% Accuracy |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 3 of 3 |
| 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

- [Rounds to 20.0% Accuracy](<https://www.sota2.com/research/sota/image-classification-on-emnist-val.md?metric=Rounds%20to%2020.0%25%20Accuracy>): Lower is better; unit: SCALAR.

- [Rounds to 33.0% Accuracy](<https://www.sota2.com/research/sota/image-classification-on-emnist-val.md?metric=Rounds%20to%2033.0%25%20Accuracy>): Lower is better; unit: SCALAR.

- [Avg Acc Post-33.0%](<https://www.sota2.com/research/sota/image-classification-on-emnist-val.md?metric=Avg%20Acc%20Post-33.0%25>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Rounds to 20.0% Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FedZMG | 15 | Not recorded | 699bdefe09c33900fcd804a3 |
| FedAvg | 15 | Not recorded | 699bdefe09c33900fcd804a3 |
| FedAdam | 35 | Not recorded | 699bdefe09c33900fcd804a3 |

### FedZMG

Rounds to 20.0% Accuracy: 15. Result ID: 699d0af55698665ec4aafd69.

[Reporting paper](<https://www.sota2.com/research/paper/699bdefe09c33900fcd804a3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699bdefe09c33900fcd804a3.md>)

Recorded evaluation configuration: {"Local Epochs":"4","Cohort Size":"10","Momentum":"0.9","Weight Decay":"0.0005"}. This record does not assert independent reproduction.

### FedAvg

Rounds to 20.0% Accuracy: 15. Result ID: 699d0af55698665ec4aafd6a.

[Reporting paper](<https://www.sota2.com/research/paper/699bdefe09c33900fcd804a3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa57171e209131ce9a25dd.md>)

Recorded evaluation configuration: {"Local Epochs":"4","Cohort Size":"10"}. This record does not assert independent reproduction.

### FedAdam

Rounds to 20.0% Accuracy: 35. Result ID: 699d0af55698665ec4aafd6b.

[Reporting paper](<https://www.sota2.com/research/paper/699bdefe09c33900fcd804a3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa58711e209131ce9ae27a.md>)

Recorded evaluation configuration: {"Local Epochs":"4","Cohort Size":"10","beta1":"0.9","beta2":"0.99","epsilon":"0.001"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
