# Pooling (machine learning)

> Reducing a matrix (or matrixes) created by an earlier convolutionary layer to a smaller matrix. Pooling usually involves taking either the maximum or average value across the pooled area.[^1]

Reducing a matrix (or matrixes) created by an earlier [convolutionary layer](https://wiki.g15e.com/pages/Convolutional%20layer.txt) to a smaller matrix. Pooling usually involves taking either the maximum or average value across the pooled area.[^1]

Pooling for vision applications is known more formally as **spatial pooling**. Time-series applications usually refer to pooling as **temporal pooling**. Less formally, pooling is often called **subsampling** or [downsampling](https://wiki.g15e.com/pages/Downsampling%20(machine%20learning.txt)).[^1]

## Footnotes

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