← Deep Learning learning pathDeep Learning · Module 2 of 10
Neurons and neural networks
Understand how artificial neurons and layers work.
📖 Key ideas
- A neuron computes a weighted sum of inputs plus a bias.
- An activation function adds non-linearity.
- Layers: input, hidden, output.
- Weights are learned from data.
- Forward pass: inputs flow to outputs to make a prediction.
💡 Real-life example: A neuron is a judge weighing several opinions (inputs) and deciding how strongly to respond.
WeightImportance of an input
BiasOffset added to the sum
Activation functionAdds non-linearity
Forward passComputing output from input