Activation Functions
CReLU
Introduced by Shang et al. in Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear UnitsCReLU, or Concatenated Rectified Linear Units, is a type of activation function which preserves both positive and negative phase information while enforcing non-saturated non-linearity. We compute by concatenating the layer output $h$ as:
$$ \left[\text{ReLU}\left(h\right), \text{ReLU}\left(-h\right)\right] $$
Source: Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 2 | 11.76% |
| Image Generation | 2 | 11.76% |
| Classification | 1 | 5.88% |
| Denoising | 1 | 5.88% |
| Fine-Grained Image Classification | 1 | 5.88% |
| Super-Resolution | 1 | 5.88% |
| Image Enhancement | 1 | 5.88% |
| Low-Light Image Enhancement | 1 | 5.88% |
| Saliency Detection | 1 | 5.88% |
Leaky ReLU
ReLU
