Bias: Difference between revisions

In a [[neural network]], [[bias]] is an additional [[input]] value that is added to the weighted sum of the input values in each neuron, before the [[activation function]] is applied. It provides the network with the ability to adjust the output of the neuron independent of the input.

The bias term is learned during the training process, along with the [[weight]]s. The bias value is adjusted to minimize the error between the predicted output and the actual output. The presence of the bias term allows the neural network to model more complex relationships between the input and output.