# Image Classification on MNIST Label Noise 0.2 (test) Source: [Image Classification on MNIST Label Noise 0.2 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:51:50.616Z Evaluation of Image Classification performance on the test split of the MNIST dataset corrupted with 20% 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 Label Noise 0.2 (test) | | Metric | Mean Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Mean Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Mean Accuracy | Result date | Report paper | | --- | --- | --- | --- | | GIW | 94.6 | Not recorded | 68fa71131e209131cea4f1d5 | | GIW | 91.44 | Not recorded | 68fa71131e209131cea4f1d5 | | Reweight | 89.28 | Not recorded | 68fa71131e209131cea4f1d5 | | CCSA | 88.06 | Not recorded | 68fa71131e209131cea4f1d5 | | Pretrain-val | 86.34 | Not recorded | 68fa71131e209131cea4f1d5 | | DANN | 85.32 | Not recorded | 68fa71131e209131cea4f1d5 | | Pretrain-val | 85.08 | Not recorded | 68fa71131e209131cea4f1d5 | | R-DIW | 82.7 | Not recorded | 68fa71131e209131cea4f1d5 | | MW-Net | 82.29 | Not recorded | 68fa71131e209131cea4f1d5 | | DIW | 80.12 | Not recorded | 68fa71131e209131cea4f1d5 | | Val-only | 79.86 | Not recorded | 68fa71131e209131cea4f1d5 | | Val-only | 78.94 | Not recorded | 68fa71131e209131cea4f1d5 | ### GIW Mean Accuracy: 94.6. Result ID: 69917c183a6c314849087581. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### GIW Mean Accuracy: 91.44. Result ID: 69917c183a6c3148490875d3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Shift Scenario":"Case (iv)"}. This record does not assert independent reproduction. ### Reweight Mean Accuracy: 89.28. Result ID: 69917c183a6c31484908757d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### CCSA Mean Accuracy: 88.06. Result ID: 69917c183a6c3148490875d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Shift Scenario":"Case (iv)"}. This record does not assert independent reproduction. ### Pretrain-val Mean Accuracy: 86.34. Result ID: 69917c183a6c31484908757c. [Reporting paper]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### DANN Mean Accuracy: 85.32. Result ID: 69917c183a6c3148490875d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Shift Scenario":"Case (iv)"}. This record does not assert independent reproduction. ### Pretrain-val Mean Accuracy: 85.08. Result ID: 69917c183a6c3148490875d0. [Reporting paper]() Recorded evaluation configuration: {"Shift Scenario":"Case (iv)"}. This record does not assert independent reproduction. ### R-DIW Mean Accuracy: 82.7. Result ID: 69917c183a6c31484908757f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### MW-Net Mean Accuracy: 82.29. Result ID: 69917c183a6c31484908757e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### DIW Mean Accuracy: 80.12. Result ID: 69917c183a6c314849087580. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. ### Val-only Mean Accuracy: 79.86. Result ID: 69917c183a6c3148490875cf. [Reporting paper]() Recorded evaluation configuration: {"Shift Scenario":"Case (iv)"}. This record does not assert independent reproduction. ### Val-only Mean Accuracy: 78.94. Result ID: 69917c183a6c31484908757b. [Reporting paper]() Recorded evaluation configuration: {"Shift type":"Label Noise","Shift intensity":0.2,"Case":"iii","Trials":5}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.