What's the definition of retraining?

In transfer learning, we always use new data to retrain the pre-trained model. But, what is the specific and official definition of retraining? Or what papers mentioned this definition, in transfer learning field or reinforcement learning field?

Topic transfer-learning reinforcement-learning

Category Data Science


There is no specific and official definition of retraining. Like most machine learning terms, the term retraining is used to describe different processes.

All the definitions of retraining share these common elements:

  1. A Model was trained on data.
  2. There is new data.
  3. Model parameters are updated with the new data.

Sometimes the old parameters are discarded. Other times the old parameters are used as priors. Sometimes both the old and the new data are used to update the parameters. Other times just the new data is used.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.