# Image Classification on MNIST 50% symmetric noise (test)

Source: [Image Classification on MNIST 50% symmetric noise (test)](<https://www.sota2.com/research/sota/image-classification-on-mnist-50-percent-symmetric-noise-test>)

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

Data updated: 2026-02-26T05:34:28.461Z

Evaluation of Image Classification performance on the test split of the MNIST dataset corrupted specifically with 50% symmetric label noise.

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) | MNIST 50% symmetric noise (test) |
| Metric | Accuracy (Test) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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 (Test)](<https://www.sota2.com/research/sota/image-classification-on-mnist-50-percent-symmetric-noise-test.md?metric=Accuracy%20(Test)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Test) | Result date | Report paper |
| --- | --- | --- | --- |
| BARE | 94.38 | Not recorded | 68fa6e271e209131cea3e2e0 |
| COT+ | 93.17 | Not recorded | 68fa6e271e209131cea3e2e0 |
| CL | 92 | Not recorded | 68fa6e271e209131cea3e2e0 |
| COT | 90.8 | Not recorded | 68fa6e271e209131cea3e2e0 |
| MR | 90.39 | Not recorded | 68fa6e271e209131cea3e2e0 |
| MN | 74.94 | Not recorded | 68fa6e271e209131cea3e2e0 |
| CCE | 74.3 | Not recorded | 68fa6e271e209131cea3e2e0 |

### BARE

Accuracy (Test): 94.38. Result ID: 699007193a6c314849fd9386.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Implementation](<https://github.com/dbp1994/masters_thesis_codes/tree/main/BARE>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### COT+

Accuracy (Test): 93.17. Result ID: 699007193a6c314849fd9381.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65bb1e209131cea0bed4.md>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### CL

Accuracy (Test): 92. Result ID: 699007193a6c314849fd9384.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa671d1e209131cea145d8.md>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### COT

Accuracy (Test): 90.8. Result ID: 699007193a6c314849fd9380.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64fe1e209131cea07aa4.md>)

[Implementation](<https://github.com/bhanML/Co-teaching>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### MR

Accuracy (Test): 90.39. Result ID: 699007193a6c314849fd9382.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65be1e209131cea0c1af.md>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### MN

Accuracy (Test): 74.94. Result ID: 699007193a6c314849fd9383.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66791e209131cea1067e.md>)

[Implementation](<https://github.com/xjtushujun/meta-weight-net>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

### CCE

Accuracy (Test): 74.3. Result ID: 699007193a6c314849fd9385.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e271e209131cea3e2e0.md>)

Recorded evaluation configuration: {"Backbone":"1-hidden layer MLP","Noise Type":"symmetric","Noise Rate (η)":0.5}. This record does not assert independent reproduction.

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