3
$\begingroup$

I'm just starting with neural networks wherein this towards data science article mentions that bias neuron is always initialized to 1. My question is why is the bias neuron in Neural networks is always initialized to 1 ? Can we initialize it to other positive and negative values say 2 or 3 or maybe 100 or -200 ? How will the output and training of the neural network get affected if we initialize with other values apart from 1

$\endgroup$
1
  • $\begingroup$ yes you can initialise it to anything. However, particular weight initialisation implementations might assume it being one. $\endgroup$
    – seanv507
    Sep 6, 2021 at 10:09

1 Answer 1

3
$\begingroup$

Yes, we can. there are many ways to initialize the weights for neural network.

In fact, I do not agree with "bias neuron is always initialized to 1".

Search neural network weight initialization in google scholar to get more information.

$\endgroup$
1
  • 3
    $\begingroup$ I think he just means the hypothetical "neuron" that stands in for a source, as in $y = W \cdot x + 1 \cdot b$ $\endgroup$
    – Neil G
    Sep 6, 2021 at 10:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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