Question

How can a single learning algorithm show avoidable bias, variance, and data mismatch at the same time?

A model can underfit the training set, show a large gap between training and dev performance, and still struggle because the dev data comes from a different distribution than the training data.

    0

    1

    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