learn deep. ship safe.Practice testsFollow ByteMunSubscribe

Technique

Regularization

Penalising model complexity (L1, L2, dropout) to reduce overfitting.

Concept map
Regularization

What should you learn before Regularization?

  1. Supervised learningLearning from labelled examples to predict a label or value.
  2. Model evaluation and metricsMeasuring how well a model works: train/test splits, accuracy, precision, recall, F1 and more.
  3. Bias–variance trade-offThe balance between a model that is too simple and one that is too sensitive to its training data.
  4. OverfittingA model learning noise in the training data, so it does well in training and badly on new data.

Connections to other concepts

Helps prevent
Overfitting

Practise Regularization

All topics