Kaiming Initialization, or He Initialization, is an initialization method for neural networks that takes into account the non-linearity of activation functions, such as ReLU activations.
A proper initialization method should avoid reducing or magnifying the magnitudes of input signals exponentially. Using a derivation they work out that the condition to stop this happening is:
$$\frac{1}{2}n_{l}\text{Var}\left[w_{l}\right] = 1 $$
This implies an initialization scheme of:
$$ w_{l} \sim \mathcal{N}\left(0, 2/n_{l}\right)$$
That is, a zero-centered Gaussian with standard deviation of $\sqrt{2/{n}_{l}}$ (variance shown in equation above). Biases are initialized at $0$.
Source: Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet ClassificationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Self-Supervised Learning | 47 | 7.36% |
Image Classification | 45 | 7.04% |
Semantic Segmentation | 35 | 5.48% |
Classification | 28 | 4.38% |
Image Segmentation | 14 | 2.19% |
Object Detection | 13 | 2.03% |
Decoder | 11 | 1.72% |
Quantization | 11 | 1.72% |
Speaker Verification | 7 | 1.10% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |