Sigmoid Activations are a type of activation function for neural networks:
$$f\left(x\right) = \frac{1}{\left(1+\exp\left(-x\right)\right)}$$
Some drawbacks of this activation that have been noted in the literature are: sharp damp gradients during backpropagation from deeper hidden layers to inputs, gradient saturation, and slow convergence.
Paper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Language Modelling | 21 | 3.01% |
Classification | 19 | 2.72% |
Sentence | 18 | 2.58% |
Image Classification | 16 | 2.29% |
Image Generation | 16 | 2.29% |
Image-to-Image Translation | 16 | 2.29% |
Management | 15 | 2.15% |
Decision Making | 14 | 2.01% |
Translation | 14 | 2.01% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |