# Image Classification on FMNIST

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

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

Data updated: 2026-02-26T05:34:30.462Z

Evaluation of Image Classification performance on the Fashion MNIST dataset, utilizing specialized accuracy metrics defined by alpha thresholds (e.g., alpha=0.1, 0.3, 0.5).

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) | FMNIST |
| Metric | Accuracy (alpha=0.1) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Accuracy (alpha=0.1)](<https://www.sota2.com/research/sota/image-classification-on-fmnist.md?metric=Accuracy%20(alpha%3D0.1)>): Higher is better; unit: PERCENT.

- [Accuracy (alpha=0.3)](<https://www.sota2.com/research/sota/image-classification-on-fmnist.md?metric=Accuracy%20(alpha%3D0.3)>): Higher is better; unit: PERCENT.

- [Accuracy (alpha=0.5)](<https://www.sota2.com/research/sota/image-classification-on-fmnist.md?metric=Accuracy%20(alpha%3D0.5)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (alpha=0.1) | Result date | Report paper |
| --- | --- | --- | --- |
| FuseFL | 83.2 | Not recorded | 68fa75721e209131cea67fa7 |
| FuseFL | 83.15 | Not recorded | 68fa75721e209131cea67fa7 |
| FuseFL | 83.05 | Not recorded | 68fa75721e209131cea67fa7 |
| Ensemble | 67.71 | Not recorded | 68fa75721e209131cea67fa7 |
| DENSE | 50.29 | Not recorded | 68fa75721e209131cea67fa7 |
| Fed-ADI | 48.49 | Not recorded | 68fa75721e209131cea67fa7 |
| Fed-DAFL | 47.14 | Not recorded | 68fa75721e209131cea67fa7 |
| FedDF | 43.58 | Not recorded | 68fa75721e209131cea67fa7 |
| FedAvg | 41.69 | Not recorded | 68fa75721e209131cea67fa7 |

### FuseFL

Accuracy (alpha=0.1): 83.2. Result ID: 6993bf143a6c314849183c88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Implementation](<https://github.com/wizard1203/FuseFL>)

Recorded evaluation configuration: {"K (number of splitted blocks)":8,"Communication rounds":8}. This record does not assert independent reproduction.

### FuseFL

Accuracy (alpha=0.1): 83.15. Result ID: 6993bf143a6c314849183c86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Implementation](<https://github.com/wizard1203/FuseFL>)

Recorded evaluation configuration: {"K (number of splitted blocks)":2,"Communication rounds":2}. This record does not assert independent reproduction.

### FuseFL

Accuracy (alpha=0.1): 83.05. Result ID: 6993bf143a6c314849183c87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Implementation](<https://github.com/wizard1203/FuseFL>)

Recorded evaluation configuration: {"K (number of splitted blocks)":4,"Communication rounds":4}. This record does not assert independent reproduction.

### Ensemble

Accuracy (alpha=0.1): 67.71. Result ID: 6993bf143a6c314849183c85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65721e209131cea0a7f3.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

### DENSE

Accuracy (alpha=0.1): 50.29. Result ID: 6993bf143a6c314849183c84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e031e209131cea3d6b6.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

### Fed-ADI

Accuracy (alpha=0.1): 48.49. Result ID: 6993bf143a6c314849183c83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67b41e209131cea17f08.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

### Fed-DAFL

Accuracy (alpha=0.1): 47.14. Result ID: 6993bf143a6c314849183c82.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66e71e209131cea130f9.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

### FedDF

Accuracy (alpha=0.1): 43.58. Result ID: 6993bf143a6c314849183c81.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa699e1e209131cea23212.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

### FedAvg

Accuracy (alpha=0.1): 41.69. Result ID: 6993bf143a6c314849183c80.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75721e209131cea67fa7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e761e209131cea40072.md>)

Recorded evaluation configuration: {"Communication rounds":1}. This record does not assert independent reproduction.

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