Glossary term
Glossary term
Architecture
A linear model that typically has many sparse input features. We refer to it as "wide" since such a model is a special type of neural network with a large number of inputs that connect directly to the output node. Wide models are often easier to debug and inspect than deep models. Although wide models cannot express nonlinearities through hidden layers, wide models can use transformations such as feature crossing and bucketization to model nonlinearities in different ways.
Contrast with deep model.
Created for this library
A search-quality team uses a wide model on memorization-heavy features like query-document IDs alongside a deep model on dense features.
An ad-tech team uses a wide model on cross-features and a deep model on dense features in its production ensemble.
A retail recommender team uses a wide-and-deep architecture so memorization of common patterns and generalization to new ones happen together.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License