Bias Term
/ˈbaɪəs tɜːrm/
An additional learnable parameter in neural networks that allows shifting the activation function.
In italiano: Termine di BiasBias terms enable neurons to activate even when all inputs are zero. Each neuron typically has one bias. They increase model flexibility and are crucial for fitting non-zero-centered data.
Examples
- Linear transformation: y = Wx + b
- Bias in each layer
- Offset parameter