# Linear model

> A model that assigns one weight per feature to make predictions. (Linear models also incorporate a bias.) In contrast, the relationship of features to predictions in deep model is generally **nonlinear**.[^1]

A [model](https://wiki.g15e.com/pages/Model%20(machine%20learning.txt)) that assigns one [weight](https://wiki.g15e.com/pages/Weight%20(machine%20learning.txt)) per [feature](https://wiki.g15e.com/pages/Feature%20(machine%20learning.txt)) to make [predictions](https://wiki.g15e.com/pages/Prediction%20(machine%20learning.txt)). (Linear models also incorporate a [bias](https://wiki.g15e.com/pages/Bias%20(machine%20learning.txt)).) In contrast, the relationship of features to predictions in [deep model](https://wiki.g15e.com/pages/Deep%20neural%20network.txt) is generally **nonlinear**.[^1]

Linear models are usually easier to train and more [interpretable](https://wiki.g15e.com/pages/Interpretability%20(machine%20learning.txt)) than deep models. However, deep models can learn complex relationships *between* features.

[Linear regression](https://wiki.g15e.com/pages/Linear%20regression.txt) and [logistic regression](https://wiki.g15e.com/pages/Logistic%20regression.txt) are two types of linear models.

## Footnotes

[^1]: https://developers.google.com/machine-learning/glossary#linear_model