Generative Models
RealNVP
Introduced by Dinh et al. in Density estimation using Real NVPRealNVP is a generative model that utilises real-valued non-volume preserving (real NVP) transformations for density estimation. The model can perform efficient and exact inference, sampling and log-density estimation of data points.
Source: Density estimation using Real NVP
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Anomaly Detection | 5 | 26.32% |
| BIG-bench Machine Learning | 2 | 10.53% |
| Defect Detection | 2 | 10.53% |
| Image Classification | 1 | 5.26% |
| Sign Language Recognition | 1 | 5.26% |
| Time Series Analysis | 1 | 5.26% |
| Object Detection | 1 | 5.26% |
| Disentanglement | 1 | 5.26% |
| Adversarial Robustness | 1 | 5.26% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
Affine Coupling
|
Bijective Transformation | |
Batch Normalization
|
Normalization | |
Normalizing Flows
|
Distribution Approximation |
