-1

I know we can't use perceptron learning algorithm to implement XOR gate because it is a lineraly inseparable problem. So my question is which learning algorithm and which neural network can we use to implement XOR gate? I tried using Delta rule, but it is not producing desired weight matrix.

Thank You!

Vipul Tyagi
  • 547
  • 3
  • 11
  • 29

1 Answers1

1

A 2 layered MLP (multi-layer perceptron) will do the trick.

Consider this article.

By the way, Wikipedia reads:

The delta rule is a gradient descent learning rule for updating the weights of the inputs to artificial neurons in a single-layer neural network.

The "single-layer neural network" here is the issue. As you said, a simple (single layer) perceptron does not have the representational power to capture XOR.

Rocco Fortuna
  • 300
  • 1
  • 11
  • Thanks for the answer. Btw, do you have any resource that contains a numerical example of delta rule. Asking a new question for this doesn't look appropriate. – Vipul Tyagi Feb 13 '21 at 14:19
  • Do you mean MADALINE can be used? – Vipul Tyagi Feb 13 '21 at 14:19
  • @VIPULTYAGI MADALINE seems to be a specific architecture of MLP that uses ADALINE units. I guess it could work? What I meant is using multiple layers in general: instead of a single layered perceptron, introduce hidden layers to obtain an MLP. – Rocco Fortuna Feb 13 '21 at 20:45
  • https://www.youtube.com/watch?v=wvXgjETpDpk this video features a numerical example of the delta rule. P.S.: please consider accepting/upvoting my answer, thanks :) – Rocco Fortuna Feb 13 '21 at 20:46