regularizer
- noun
- /ˈrɛɡjʊləˌraɪzər/
- Specialized
:
regularizador
- In order to reduce overfitting, the data scientist implemented a regularizer in the training process.
Examples
-
Depending on the field, deep learning is usually competing with things like hidden Markov models, Gaussian processes, mixture models, or various fancy regression approaches like boosting and sparse regularizers.
Blog text (10) -
An even better approach is the explicit use of gd0 in the Mumford-Shah regularizer concerned by Storath and co-workers 25.
Academic text (2016) -
The use of a proper regularizer can significantly improve the performance of machine learning models.
-
By adding a regularizer to the model, researchers can constrain the weight values and enhance generalization.
Synonyms
A method used in machine learning to keep a model simple so it predicts new data better
Surface Forms
Morphology
regularizer = regularize (semi-transparent) = regular + ize + er
Formed by regular + -ize + -er (predictable derivation), but the specific technical meaning as an ML 'method or tool' is specialized and not obvious to general learners.
Etymology
Regularizer comes from regular, which in turn comes from regula meaning 'rule'. In machine learning, a regularizer is like a simple 'rule' added to a model to stop it from memorizing the training data, so it learns patterns that work on new examples.
