Why is a neural bias system necessary for backpropagating a neural network that the XOR operator recognizes?

Question

I posted yesterday a question about the problems that I had with my backpropagating neural network for the XOR operator. I did a little more work and realized that this could be due to the lack of neural bias.

My question is, what is the role of the neural syndrome as a whole and what is its role in the backpropagating neural network that the XOR operator recognizes? Is it possible to create one without neural bias?

Answer 1

You can create a neural network without a neural system ... this will work fine, but for more information, I would recommend that you see the answers to this question:

The role of bias in neural networks

Update: The role of a neural nerve in a neural network that is trying to "hit" the XOR model is to minimize the size of the neural network. Usually for a "primitive" (not sure if this is the right term) logic functions, such as AND , OR , NAND , etc., you try to create a neural network with 2 input neurons, 2 hidden neurons and 1 output neuron. This is not possible for XOR , because the easiest way to simulate XOR is with two NAND s:

You can consider A and B as your input neurons, the gate in the middle is your “bias” neuron, the next two gates are your “hidden” neurons, and finally you have an output neuron. You can solve XOR without a neural bias, but for this you will need to increase the number of hidden neurons to a minimum of 3 hidden neurons. In this case, the 3rd neuron essentially acts as a neural syndrome. Here is another question that a neuron with bias discusses regarding the XOR : XOR problem , solvable with a 2x2x1 neural network without bias?