Concept icon
Concept

A Model Can Show Both Bias and Variance

A classifier with an estimated bias of 14% and variance of 16% has both kinds of error. It does poorly on the training set and becomes even less reliable on the dev set. In this situation, the usual labels of underfitting and overfitting are both partly true, so they do not describe the model very cleanly.

0

1

Concept icon
Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Machine Learning Strategy

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Yearning @ DeepLearning.AI

Related