# Machine Unlearning on TinyImageNet (30% Random Forget) Source: [Machine Unlearning on TinyImageNet (30% Random Forget)]() Collection index: [llms.txt]() Data updated: 2026-04-13T09:47:26.887Z Evaluation of Machine Unlearning performance on TinyImageNet, specifically configured to forget 30% of the training data via random selection. Metrics focus on forgetting rates (Dr, Df, Dtest) and Membership Inference Attack (MIA) success. 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 | Machine Unlearning | | Dataset (including any stated split) | TinyImageNet 30% random data forgetting | | Metric | Dr (Train Set) | | 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 - [Dr (Train Set)](): Higher is better; unit: PERCENT. - [Df (Train Set)](): Higher is better; unit: PERCENT. - [Dtest (Test Set)](): Higher is better; unit: PERCENT. - [MIA Success Rate](): Lower is better; unit: PERCENT. - [Average Gap](): Lower is better; unit: PERCENT. ## Recorded results | Method | Dr (Train Set) | Result date | Report paper | | --- | --- | --- | --- | | IEU w/GA | 68.7 | Not recorded | 69dc789f7d0172140ec78bd0 | | FT | 68.4 | Not recorded | 69dc789f7d0172140ec78bd0 | | IEU w/Noisy | 65.9 | Not recorded | 69dc789f7d0172140ec78bd0 | | IEU w/GA+Noisy | 65.6 | Not recorded | 69dc789f7d0172140ec78bd0 | | Retraining | 62.3 | Not recorded | 69dc789f7d0172140ec78bd0 | | SCRUB | 53 | Not recorded | 69dc789f7d0172140ec78bd0 | | RL | 51.6 | Not recorded | 69dc789f7d0172140ec78bd0 | | SALUN | 34.8 | Not recorded | 69dc789f7d0172140ec78bd0 | ### IEU w/GA Dr (Train Set): 68.7. Result ID: 69dcbb9847f34ce924c411c3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### FT Dr (Train Set): 68.4. Result ID: 69dcbb9847f34ce924c411bf. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### IEU w/Noisy Dr (Train Set): 65.9. Result ID: 69dcbb9847f34ce924c411c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### IEU w/GA+Noisy Dr (Train Set): 65.6. Result ID: 69dcbb9847f34ce924c411c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Retraining Dr (Train Set): 62.3. Result ID: 69dcbb9847f34ce924c411be. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### SCRUB Dr (Train Set): 53. Result ID: 69dcbb9847f34ce924c411c1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### RL Dr (Train Set): 51.6. Result ID: 69dcbb9847f34ce924c411c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### SALUN Dr (Train Set): 34.8. Result ID: 69dcbb9847f34ce924c411c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.