I am trying to learn how to manipulate McCulloch-Pitts neurons in order to determine their weights and bias based off of inputs. In this example I have inputs:

x, y, z ∈ {−1,1}

The neuron's output is z if x = -1 and y = 1, and is -1 otherwise.

How do I compute the weights and bias for the neuron? An example or link to an example would be greatly appreciated. (Having a hard time finding examples online). Thanks!

So far I have been using the equation out = sgn((w1)(in1) + (w2)(in2) + (w3)(in3) - theta), but I am unsure whether or not I am taking the correct initial steps to solving this problem.


The weights and bias for a given neuron depend on the task you are trying to solve. Each computation will have a different set of weights that will implement it.

  • $\begingroup$ Does the method for finding those weights (which I think the question is asking for) also vary by task? Why? $\endgroup$ – Christian Hummeluhr May 15 '15 at 7:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.