● mastered · kind concept · level 3 · 10h
- Requiere: Neural Network
The non-linearities — ReLU, sigmoid, tanh, GELU — that let stacked layers represent more than a single linear map. Their shape governs gradient flow.
In artificial neural networks, the activation function of a node is a function that calculates the output of the node based on its individual inputs and their weights. Nontrivial problems can be solved using only a few nodes if the activation function is nonlinear.
Enlaces
- Requiere: Neural Network