Normalization
Activation Normalization
Introduced by Kingma et al. in Glow: Generative Flow with Invertible 1x1 ConvolutionsActivation Normalization is a type of normalization used for flow-based generative models; specifically it was introduced in the GLOW architecture. An ActNorm layer performs an affine transformation of the activations using a scale and bias parameter per channel, similar to batch normalization. These parameters are initialized such that the post-actnorm activations per-channel have zero mean and unit variance given an initial minibatch of data. This is a form of data dependent initilization. After initialization, the scale and bias are treated as regular trainable parameters that are independent of the data.
Source: Glow: Generative Flow with Invertible 1x1 Convolutions
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Generation | 3 | 9.38% |
| Anomaly Detection | 2 | 6.25% |
| OOD Detection | 2 | 6.25% |
| Dimensionality Reduction | 2 | 6.25% |
| Disentanglement | 1 | 3.13% |
| Unsupervised Anomaly Detection | 1 | 3.13% |
| Text-To-Speech Synthesis | 1 | 3.13% |
| Word Alignment | 1 | 3.13% |
| Emotion Classification | 1 | 3.13% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
