AI Blog

AI Blog

by Michele Laurelli

Back to Glossary

Inception Module

Architecture
Definition

CNN building block applying multiple filter sizes in parallel and concatenating results.

Captures features at different scales simultaneously. Uses 1x1, 3x3, 5x5 convolutions plus pooling. Reduces parameters with 1x1 bottleneck convolutions.

Examples

1

GoogLeNet/Inception v1

2

Multi-scale feature extraction

3

Efficient CNN design

Related Terms

CNN (Convolutional Neural Network)

A deep learning architecture specialized for processing grid-like data such as images, using convolutional layers.

Michele Laurelli - AI Research & Engineering