Perceptron learning, Delta learning and LMS learning are learning methods…
2014
Perceptron learning, Delta learning and LMS learning are learning methods which falls under the category of
- A.
Error correction learning – learning with a teacher
- B.
Reinforcement learning – learning with a critic
- C.
Hebbian learning
- D.
Competitive learning – learning without a teacher
Attempted by 62 students.
Show answer & explanation
Correct answer: A
Correct classification: Perceptron learning, Delta learning and LMS learning belong to error-correction supervised learning (learning with a teacher).
Key points:
Supervision: A teacher provides desired outputs; the algorithm uses the difference between actual and desired outputs (the error) to update weights.
Perceptron learning: Used for binary classification with a hard threshold activation. Weights are adjusted when the prediction is incorrect (update driven by the sign of the error).
Delta rule (Widrow–Hoff) and LMS: These perform gradient-descent on the mean squared error. The typical update is proportional to (desired output − actual output) times the input, which directly reduces squared error.
Difference between perceptron and delta/LMS: Perceptron uses a non-differentiable threshold and updates based on misclassification; delta/LMS assume differentiable responses and minimize squared error continuously.
Therefore these methods are categorized as error-correction (supervised) learning — learning with a teacher.