Regularization
Probability Guided Maxout
A regularization criterion that, differently from dropout and its variants, is deterministic rather than random. It grounds on the empirical evidence that feature descriptors with larger L2-norm and highly-active nodes are strongly correlated to confident class predictions. Thus, the criterion guides towards dropping a percentage of the most active nodes of the descriptors, proportionally to the estimated class probability
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Decision Making | 4 | 8.33% |
| Visual Reasoning | 3 | 6.25% |
| Semantic Segmentation | 3 | 6.25% |
| Time Series Analysis | 2 | 4.17% |
| Continual Learning | 1 | 2.08% |
| Incremental Learning | 1 | 2.08% |
| Benchmarking | 1 | 2.08% |
| GPT-4 | 1 | 2.08% |
| Language Modelling | 1 | 2.08% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
