0
$\begingroup$

If a neuron uses a non linear activation function such as a sigmoid function, then the output of that neuron can be any value between 0 and 1. suppose if the activation function results in value like 0.6, what will be the output of that neuron?

Should it be 0.6 or 1?

$\endgroup$
1
$\begingroup$

It depends on the architecture. As Andy said, normally the activation is equivalent to the output value. I've seen more complex designs, however, where activations are calculated and then must pass one or more thresholds, which trigger one or more predefined outputs (such as a Boolean value, a software function, or some other type of response that isn't exactly equivalent to the activation, or even to its data type). It is also possible in more exotic nets to have multiple activations in a single neuron. Recently, I found it much easier to model the simple and complex dual cells Fukushima uses in cognitrons and neocognitrons in such a non-standard way. I hope that helps. :)

$\endgroup$
0
$\begingroup$

If the activation function results in an output of 0.6 then the output of that neuron is 0.6.

The activation function is what determines the output of the neuron.

$\endgroup$

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.