Lecture 13 - Neural Networks Part 1

Question 1

What are the components of a McCulloch-Pitts Neuron?

Answer 1

Inputs - with weights

Threshold (critical value)

Question 2

How do McCulloch-Pitts Neurons work?

Answer 2

Weighted summation of all inputs (sum of all inputs multiplied by their weights)

Output is either 0 or 1 depending on whether sum value is greater or less than the threshold

Question 3

What boolean logic functions can a MP Neuron compute?

Answer 3

AND

OR

NOT

Question 4

Why can XOR not be computed with one MP Neuron?

Answer 4

There is no way to separate the values of XOR using a straight line:

Question 5

What is the general equation for a hyperplane in n-dimensional space?

Answer 5

Question 6

The only functions that an n-input MP Neuron can compute are those defined by a

Answer 6

linear surface

Question 7

MP Neurons are also known as

Answer 7

perceptrons

Linear Separation Units