3.3. Statistical Learning Theory

Artificial intelligence technologies aim at mining potential disciplines from the training samples or observable data, and applying the mined disciplines to predict the testing samples or future data. Most of the current artificial intelligence techniques, including ANNs, are to find out the disciplines that the samples are subject to when the number of samples is close to be infinite. This is named as the empirical risk minimization (ERM), which aims to minimize the error on the training samples (Zhang, 2010).
The neural network based on ERM is commonly used in plenty of engineering applications. In the training process, it is hoped to build up a neural network model of the system observed, namely to deduce responses ...

Get Intelligent Fault Diagnosis and Remaining Useful Life Prediction of Rotating Machinery now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.