Fundamentals of Machine Learning for Predictive Data Analytics by John D. Kelleher;Brian Mac Namee;Aoife D'Arcy;

Author:John D. Kelleher;Brian Mac Namee;Aoife D'Arcy;
Language: eng
Format: epub
Tags: machine learning; predictive analytics; data analyticsdata science; data mining; deep learning; supervised learning; unsupervised learning; reinforcement learning; machine learning case studies;
Publisher: MIT Press

Note that neurons are often referred to as units, and they are distinguished by the type of activation function they use. Hence a neuron that uses a logistic activation function is referred to as a logistic unit, and a unit that uses the rectifier function is known as a rectified linear unit or ReLU. We will explain why the logistic function became so popular as an activation function when we introduce the backpropagation algorithm (see Section 8.3[403]), and why the rectified linear function replaced it when we explain the vanishing gradient problem (see Section 8.4.1[434]). Figure 8.2[387] shows plots of some of the activation functions that have been popular in neural networks over the last few decades, including the threshold, logistic, tanh, and rectifier linear functions.

Figure 8.2

Plots for activation functions that have been popular in the history of neural networks.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.