Difference between logistic regression and logistic neuron

Question

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")

2. They are both binary classifiers

3. Both classifiers are symmetric and give the same output for the same input.

   But what is the difference between the two?

Answer 1

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