Optimization
Adversarial Model Perturbation
Introduced by Zheng et al. in Regularizing Neural Networks via Adversarial Model PerturbationBased on the understanding that the flat local minima of the empirical risk cause the model to generalize better. Adversarial Model Perturbation (AMP) improves generalization via minimizing the AMP loss, which is obtained from the empirical risk by applying the worst norm-bounded perturbation on each point in the parameter space.
Source: Regularizing Neural Networks via Adversarial Model Perturbation
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Action Detection | 4 | 9.30% |
| Activity Detection | 4 | 9.30% |
| Denoising | 4 | 9.30% |
| Vocal Bursts Type Prediction | 2 | 4.65% |
| Quantization | 2 | 4.65% |
| Clustering | 2 | 4.65% |
| Benchmarking | 1 | 2.33% |
| Autonomous Driving | 1 | 2.33% |
| Motion Forecasting | 1 | 2.33% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
