# Image Classification on Structured Noise Fashion MNIST (train)

Source: [Image Classification on Structured Noise Fashion MNIST (train)](<https://www.sota2.com/research/sota/image-classification-on-structured-noise-fashion-mnist-train>)

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

Data updated: 2026-02-26T05:27:36.735Z

Evaluation of Image Classification performance on the training split of the Structured Noise Fashion MNIST 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 | Classification |
| Dataset (including any stated split) | Structured Noise Fashion MNIST (train) |
| Metric | Accuracy |
| 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](<https://www.sota2.com/research/sota/image-classification-on-structured-noise-fashion-mnist-train.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ERM 2 | 84.66 | Not recorded | 68fa67361e209131cea14d1e |
| ERM | 83.49 | Not recorded | 68fa67361e209131cea14d1e |
| Robust Min Max | 82.78 | Not recorded | 68fa67361e209131cea14d1e |
| ERM 1 | 81.8 | Not recorded | 68fa67361e209131cea14d1e |
| Optimal | 75 | Not recorded | 68fa67361e209131cea14d1e |
| ERM No Noise | 74.79 | Not recorded | 68fa67361e209131cea14d1e |
| IRM | 52.57 | Not recorded | 68fa67361e209131cea14d1e |
| F-IRM GAME | 51.54 | Not recorded | 68fa67361e209131cea14d1e |
| V-IRM GAME | 47.7 | Not recorded | 68fa67361e209131cea14d1e |

### ERM 2

Accuracy: 84.66. Result ID: 698d1f2e8b517eb4331f2af5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

Recorded evaluation configuration: {"Environment":"2"}. This record does not assert independent reproduction.

### ERM

Accuracy: 83.49. Result ID: 698d1f2e8b517eb4331f2af3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Robust Min Max

Accuracy: 82.78. Result ID: 698d1f2e8b517eb4331f2af6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ERM 1

Accuracy: 81.8. Result ID: 698d1f2e8b517eb4331f2af4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

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

### Optimal

Accuracy: 75. Result ID: 698d1f2e8b517eb4331f2afb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ERM No Noise

Accuracy: 74.79. Result ID: 698d1f2e8b517eb4331f2afa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

Recorded evaluation configuration: {"noise\_free":true}. This record does not assert independent reproduction.

### IRM

Accuracy: 52.57. Result ID: 698d1f2e8b517eb4331f2af9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673c1e209131cea15133.md>)

[Implementation](<https://github.com/facebookresearch/InvariantRiskMinimization>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### F-IRM GAME

Accuracy: 51.54. Result ID: 698d1f2e8b517eb4331f2af7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

[Implementation](<https://github.com/IBM/IRM-games>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### V-IRM GAME

Accuracy: 47.7. Result ID: 698d1f2e8b517eb4331f2af8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d1e.md>)

[Implementation](<https://github.com/IBM/IRM-games>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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