training phase

Terms from Artificial Intelligence: humans at the heart of algorithms

The glossary is being gradually proof checked, but currently has many typos and misspellings.

In machine learning the training (or learning) phase uses training data to create a model that is subsequently used for new situations/inputs in the application phase. In some cass there may be a separate validation phase.
The two main forms of training are:
  • supervised learning.– when the training data is labelled, that is each training data item consists of and input-output or input-class pair;
  • unsupervised learning.– when there is no labelling and the system looks for patterns or clusters in the data.

Used in Chap. 5: page 60; Chap. 6: page 76; Chap. 9: pages 117, 118

Also known as learning phase, training

Training phase of machine learning.