Weights
Explain what weights do in a simple artificial neuron.
Core explanation
A weight controls how strongly a corresponding input contributes to a neuron's weighted sum. A positive weight increases the sum when the input is positive; a negative weight would decrease it. The magnitude affects the size of the contribution.
In the spam-filter example:
- promotional language weight: 0.4
- suspicious link weight: 0.9
- untrusted sender weight: 0.7
If x2 = 1, its contribution is 1 × 0.9 = 0.9.
Formula
z = (x1×w1) + (x2×w2) + (x3×w3) + b
Weights are trainable parameters in a typical neural network. During training, the learning process adjusts them to reduce the model's loss/objective.
Important distinction
A weight is not a business priority score assigned manually in every model. In a trained neural network, weights are generally learned parameters. The illustrative values here are chosen for teaching arithmetic, not extracted from a real spam classifier.
Knowledge check
1. What does a weight control? The scale and direction of an input's contribution to the weighted sum. 2. Are the example weights (0.4, 0.9, 0.7) verified parameters from a deployed spam filter? No. They are illustrative teaching values.
Recovered from the available teaching record; not a verbatim transcript.
Introduced; learner mastery not verified.
Help improve this learning path.
Suggest a concept we should teach, or tell us what is unclear. No account is required. Email is optional and only for a reply. Do not include confidential or sensitive information.