A Theoretical Analysis of Feature Pooling in Visual RecognitionA Theoretical Analysis of Feature Pooling in Visual RecognitionBoureau, Y-Lan and Ponce, Jean and LeCun, Yann2010
Paper summarymartinthomaThis paper analyzes max pooling and average pooling, as it is used in many convolutional neural networks (CNNs).
## Why pooling is used
* invariance to image transformations
* more compact representations (- remove irrelevant information)
* better robustness to noise and clutter
## Max pooling or average pooling?
No clear answer to that. Sometimes one seems to be better, sometimes the other, sometimes something in between.
This paper analyzes max pooling and average pooling, as it is used in many convolutional neural networks (CNNs).
## Why pooling is used
* invariance to image transformations
* more compact representations (- remove irrelevant information)
* better robustness to noise and clutter
## Max pooling or average pooling?
No clear answer to that. Sometimes one seems to be better, sometimes the other, sometimes something in between.