Topic 3: Perceptron

Which of the following best describes the McCulloch-Pitts (MP) neuron model?

• It uses a weighted sum and applies a sigmoid activation function

• It outputs a continuous value between 0 and 1

• It computes a threshold function based on binary inputs

• It uses gradient descent to update weights

What is the key difference between the MP neuron and the Perceptron?

• MP neuron supports continuous inputs while Perceptron does not

• Perceptron uses learnable weights, while MP neuron uses fixed binary.

• MP neuron is used for regression tasks

• Perceptron cannot solve linearly separable problems

Which of the following is a limitation of the McCulloch-Pitts (MP) neuron model?

• It uses non-linear activation functions

• It cannot process multiple inputs

• It cannot learn from data because it has fixed weights and threshold

• It supports continuous-valued outputs

Which of the following problems can be solved by a single-layer perceptron?

• XOR logic function

• Image classification

• Linearly separable problems like AND and OR

• Recurrent sequence modeling

In the Perceptron learning rule, how are weights updated?

• w = w + η (target−predicted) × input

• w = w × η + input

• w = w − η × input

• Weights are not updated in the Perceptron model