Generalization
Distributional Generalization
Introduced by Nakkiran et al. in Distributional Generalization: A New Kind of GeneralizationDistributional Generalization is a type of generalization that roughly states that outputs of a classifier at train and test time are close as distributions, as opposed to close in just their average error. This behavior is not captured by classical generalization, which would only consider the average error and not the distribution of errors over the input domain.
Source: Distributional Generalization: A New Kind of Generalization
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Matrix Completion | 1 | 33.33% |
| Distributional Reinforcement Learning | 1 | 33.33% |
| 2D Object Detection | 1 | 33.33% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
