1
$\begingroup$

What is the difference between a logistic neuron in a neural net and a logit regression?

  1. They both follow the Sigmoid function plotted below

Here is the R Code for reasons of reproducibility

f <- function(x) {
   1 / (1 + exp(-x))
}

curve(f, xlim = c(-6,6), ylim= c(-0.5,1.5), main = "Sigmoid function", col.main = "red")

enter image description here

  1. They are both binary classifiers
  2. Both classifiers are symmetric and give the same output for the same input.

    But what is the difference between the two?

$\endgroup$
1
$\begingroup$

A neural network can be considered as a networked set of logistic regression units.

While a single logistic regression can perform as a classifier on it's own it's not suited for problems where input dimensions are very high and your data is not linearly separable.

By using multiple such units, a neural network attempts to approximate any given function. The more important aspect in using a neural network however is in knowing how to train each of these units.

Here's another question similar to yours: Difference between logistic regression and neural networks

$\endgroup$

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.