Sigmoid function

A mathematical function that “squishes” an input value into a constrained range, typically 0 to 1 or -1 to +1. That is, you can pass any number (two, a million, negative billion, whatever) to a sigmoid and the output will still be in the constrained range.¹

The sigmoid function has several uses in machine learning, including:

• Converting the raw output of a logistic regression or multinomial regression model to a probability.
• Acting as an activation function in some neural networks.

Formula

The sigmoid function over an input number $x$ has the following formula:

$$sigmoid(x) = \frac{1}{1 + e^{-x}}$$

In machine learning, x is generally a weighted sum.¹

Footnotes

1. developers.google.com/machine-learning/glossary#sigmoid-function ↩ ↩²