MCCULLOCH-

PITTS NEURON

PERCEPTRON
MODEL
 • McCulloch(neuroscientist) and Pitts(logician) proposed a highly
MCCULLOCH simplified computation model of the neuron(1943).
• X1,x2…xn are inputs
-PITTS • g aggregates the inputs and the function f takes decision based

NEURONS(19
on this aggregation
• Inputs can be excitatory or inhibitory

43) • y=0 if any xi is inhibitory else

• is called the thresholding parameter. This is called Thresholding

logic
• Represents Boolean function
Implementation of Boolean function using
Mcculloch Pitts(MP) neuron

0
3 1 0
Reprsentation of OR Boolean function
Reprsentation of AND Boolean function
For more than 2 inputs
Concept so far
What about Non-Boolean(say real)
inputs?
Perceptron
• Frank Rosenblatt, an American psychologist
proposed the classical perceptron model(1958)
• A more general computational model than
McCulloch Pitts neurons
• Main difference: Introduction to numerical weights
for inputs and mechanism for learning these
weights
• Used an algo to facilitate supervised learning for
binary classifier
• Inputs are no longer limited to Boolean values
• Refined and carefully analyzed by Minsky and
Papert(1969)-their model is referred to perceptron
model here
What kind of functions can be implemented using the
perceptron? Any difference from McCulloch Pitts neurons?