▶Linear score
The neuron aggregates inputs using weights and a bias.
Neural Networks: Basic Architecture
Weighted inputs, bias, and activation
A neuron first combines its inputs linearly, then applies an activation function. This simple computation is the building block of a neural network.
watchone idea
→manipulateone example
→leave withone intuition
The essential idea: learned weights form a score and the activation transforms that score into an output.
One Concept · One Example
What to notice
- Weights set the influence of each input.
- Bias shifts the activation threshold.
- The activation controls the output shape.
Inside one neuron
Activated output
A nonlinear function transforms the score into the neuron output.
See a perceptron on the sigmoid curve
Use the exact weights, bias, and input from the slides. The graph connects the neuron's linear input z to its sigmoid output y.
A neuron is a weighted sum plus bias passed through an activation function.