Neurons

---

aliases: - Neuron

A neuron is a unit in a layer of a Neural Network. It receives input data from the previous layer on which it performs non-linear transformation before propagating it to the next layer.

Each neuron has a vector of weights $\mathbf{w}$ and a bias $\mathbf{b}$ and performs its calculations by multiplying its input $\mathbf{x}$ with $\mathbf{w}$ and adding $\mathbf{b}$ and afterwards, apply an Activation Function $g$ to the result. Mathematically a neuron calculates $g\left(\mathbf{x}\mathbf{w}+\mathbf{b}\right)$.