# Fair Classification on CelebA level-3 snow noise (corrupted)

Source: [Fair Classification on CelebA level-3 snow noise (corrupted)](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted>)

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

Data updated: 2026-06-23T06:34:55.370Z

Evaluation of fair classification performance on the CelebA dataset corrupted specifically with level-3 snow noise, measured using fairness metrics like Acc s+, Acc s-, Delta Acc, and Delta p.

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 | Fair Classification |
| Dataset (including any stated split) | CelebA level-3 snow noise (corrupted) |
| Metric | Acc s+ |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Acc s+](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted.md?metric=Acc%20s%2B>): Higher is better; unit: PERCENT.

- [Acc s-](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted.md?metric=Acc%20s->): Higher is better; unit: PERCENT.

- [Overall Accuracy](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

- [Delta Accuracy](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted.md?metric=Delta%20Accuracy>): Lower is better; unit: PERCENT.

- [Delta p](<https://www.sota2.com/research/sota/fair-classification-on-celeba-level-3-snow-noise-corrupted.md?metric=Delta%20p>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Acc s+ | Result date | Report paper |
| --- | --- | --- | --- |
| MSAM | 85.04 | Not recorded | 68fa77be1e209131cea750f9 |
| SAM | 85 | Not recorded | 68fa77be1e209131cea750f9 |
| GroupSAM | 84.97 | Not recorded | 68fa77be1e209131cea750f9 |
| Vanilla | 84.57 | Not recorded | 68fa77be1e209131cea750f9 |
| FairSAM | 81.75 | Not recorded | 68fa77be1e209131cea750f9 |
| Reweighed | 80.91 | Not recorded | 68fa77be1e209131cea750f9 |
| GroupDRO | 79.08 | Not recorded | 68fa77be1e209131cea750f9 |
| FairReg | 63.15 | Not recorded | 68fa77be1e209131cea750f9 |

### MSAM

Acc s+: 85.04. Result ID: 6a3a2906fbc373592c33ce89.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc61e209131cea8a5c1.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### SAM

Acc s+: 85. Result ID: 6a3a2906fbc373592c33ce88.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69d81e209131cea248fb.md>)

[Implementation](<https://github.com/google-research/sam>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### GroupSAM

Acc s+: 84.97. Result ID: 6a3a2906fbc373592c33ce8a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa705b1e209131cea4a93e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### Vanilla

Acc s+: 84.57. Result ID: 6a3a2906fbc373592c33ce84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### FairSAM

Acc s+: 81.75. Result ID: 6a3a2906fbc373592c33ce8b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### Reweighed

Acc s+: 80.91. Result ID: 6a3a2906fbc373592c33ce86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### GroupDRO

Acc s+: 79.08. Result ID: 6a3a2906fbc373592c33ce87.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67561e209131cea159c7.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

### FairReg

Acc s+: 63.15. Result ID: 6a3a2906fbc373592c33ce85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77be1e209131cea750f9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63371e209131ce9fd7c7.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Target attribute":"Big Nose","Sensitive attribute":"Age","Corruption type":"level-3 snow noise"}. This record does not assert independent reproduction.

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