CNSLT ACADEMY / INSIDE A NEURON

Bias

FOUNDATIONSBEGINNERFREE TEXT LESSONVIDEO 🔒 PRO

Explain the role of bias in a simple neuron and distinguish it from an external decision threshold.

Core explanation

Bias is an additional trainable parameter added to the weighted sum. It shifts the value before the activation function is applied. It allows the neuron to produce a non-zero pre-activation value even when all input features are zero.

Formula

z = Σ(xᵢ × wᵢ) + b

In the first illustrative spam-filter example, b = -0.5. When all inputs are zero: z = 0 + (-0.5) = -0.5; sigmoid(-0.5) ≈ 0.378.

Distinguish bias from threshold

Bias is part of the neuron's learned computation. A decision threshold is a separate policy or decision rule applied to an output score, for example, deciding to label a message as spam when the score is at least 0.70.

They are not interchangeable.

Common misconception

Misconception: Bias is the same as the spam-classification threshold.

Correction: Bias shifts the neuron's pre-activation value; the threshold is an external decision rule applied to a model output.

Knowledge check

1. What does bias add to the computation? A learned offset added to the weighted sum. 2. Is the bias the same as the classification threshold? No.

LESSON STATUS

Recovered from the available teaching record; not a verbatim transcript.

Introduced; learner mastery not verified.

LEARNER FEEDBACK

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.

COURSE PROGRESS0/10 complete
Not yet completed
CHECK YOUR UNDERSTANDING