Quiz report
Answer review
Quiz configuration
165 questions are available across multiple activities.
Learning tools
Review formulas, terminology, and an interactive decision boundary.
Formula reference
Net input: z = w·x + b
Step output: ŷ = 1 when z ≥ 0; otherwise 0
Weight update: w ← w + η(y − ŷ)x
Bias update: b ← b + η(y − ŷ)
Glossary
Weights control feature influence. Bias shifts the boundary.
Two classes are linearly separable when one linear boundary can divide them without training errors.
Convergence occurs when training reaches a stable solution. For separable data, the perceptron reaches a separator under standard assumptions.