Learn Before
Concept icon
Concept

Ways to Reduce High Variance

When a model has high variance, common remedies include collecting more training examples, adding regularization, using early stopping, simplifying the feature set, or changing the model so it is less flexible. Another option is to adjust the input representation or architecture if error analysis shows that the current setup is too sensitive to the training data.

Reducing model size can also lower variance, but it should be treated carefully. If computation is not the main limitation, regularization often gives better overall classification performance than simply making the network smaller.

0

1

Concept icon
Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI