Activation Functions
Shifted Softplus
Introduced by Schütt et al. in SchNet: A continuous-filter convolutional neural network for modeling quantum interactionsShifted Softplus is an activation function ${\rm ssp}(x) = \ln( 0.5 e^{x} + 0.5 )$, which SchNet employs as non-linearity throughout the network in order to obtain a smooth potential energy surface. The shifting ensures that ${\rm ssp}(0) = 0$ and improves the convergence of the network. This activation function shows similarity to ELUs, while having infinite order of continuity.
Source: SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Formation Energy | 3 | 27.27% |
| Property Prediction | 2 | 18.18% |
| BIG-bench Machine Learning | 2 | 18.18% |
| Robust Design | 1 | 9.09% |
| 3D Pose Estimation | 1 | 9.09% |
| Drug Discovery | 1 | 9.09% |
| Total Energy | 1 | 9.09% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
