Weight (machine learning)

A value that a model multiplies by another value. Training is the process of determining a model’s ideal weights; inference is the process of using those learned weights to make predictions.¹

Footnotes

1. developers.google.com/machine-learning/glossary#weight