Glossary term
Glossary term
Foundations
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. A plot of the sigmoid activation function looks as follows:
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.
Click the icon to see the math.
Created for this library
A churn modeling team uses a sigmoid function on its output layer to produce a calibrated probability per customer.
An ad-tech team uses a sigmoid function in its click prediction model so outputs are interpretable as probabilities.
A medical screening team uses a sigmoid function on its binary classifier so clinical reviewers can interpret outputs as probabilities.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License