Global average pooling

Photo by Alina Grubnyak on Unsplash

What is it and why is it used?

  • Normally if we want to flatten the outputs from the convolutional layer’s we try to use fully connected layers but these layers require a lot of parameters and therefore computational expensive.
  • This is were global average pooling comes in… instead of using a Fully Connected layer we can design a network such that the last convolutional layer has the number of channels equal to the number of outputs and just take each of the channel and average the values into the number of output values.

Advantages:

  • It is native to the convolution structure by enforcing correspondences between feature maps and categories. Thus the feature maps can be easily interpreted as categories confidence maps.
  • There is no parameter to optimize in the global average pooling thus overfitting is avoided at this layer. Furthermore, global average pooling sums out the spatial information, thus it is more robust to spatial translations of the input.
  • FC’s don’t understand the concept of feature maps.

--

--

--

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

What a goal! Here are our tactics and how you can be a winner using Power BI Goals

Explorative Data Analysis of Covid-19 Data in Italy

Experience of Virtual Internship with LetsGrowMore as a Data Science Intern

Football Player Ranking with Classification and Streamlit App

The Fine Line Between Predictive and Prescriptive Analytics (with examples)

Stop Thinking About Data Lakes and Start Thinking About Data Gardens

Your Subjective Probability

Exploring, Clustering and Mapping Toronto’s Crimes

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Syed Rahim Saqib

Syed Rahim Saqib

More from Medium

Object Localization using Global Average Pooling Layers

A simple take on effusion detection using chest X-rays

x-ray classes comparison

Paper summary: Adversarial Examples for Semantic Segmentation and Object Detection.

Inception Network