How deeply does Machine Learning Algorithms in Depth cover Inception module?

Defines variable receptive fields and their motivation, but stops short of deriving or implementing the module.

Inception module

As explained in Machine Learning Algorithms in Depth

An inception module is a neural-network block that applies convolutions with multiple kernel sizes to obtain features at different receptive-field scales.

An inception module is a convolutional neural-network block that applies different kernel sizes to create variable receptive fields. The resulting feature-map stack can capture sparse correlation patterns at multiple scales. section 11.5

Its role in convolutional architecture follows three linked ideas:

Different kernel sizes expose features at different receptive-field scales. section 11.5
The resulting feature-map stack captures sparse correlation patterns. section 11.5
GoogLeNet uses inception modules to achieve high ImageNet accuracy with fewer parameters than AlexNet or VGG. section 11.5

GoogLeNet in the CNN progression

Inception modules mark a progression from the convolution, ReLU, and max-pooling operations of LeNet toward GoogLeNet’s multi-scale convolutional design; later architectures introduce residual connections or image-patch transformers. section 11.5

How Machine Learning Algorithms in Depth covers it

Cover of Machine Learning Algorithms in Depth

liveBook manning.com

Read Machine Learning Algorithms in Depth on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Math and Architectures of Deep LearningA multi-branch convolutional network block that increases a network's depth and width efficiently.

Compare how 2 books explain Inception module →

Related concepts

GoogleNetused by · GoogLeNet is a deep convolutional neural network architecture that uses inception modules for image recognition.
pointwise convolutionuses · A pointwise convolution is a convolution with a 1×1 spatial kernel that mixes or transforms channels independently at each spatial position.
Inceptionimplemented by · A family of convolutional neural network architectures that use parallel multi-scale branches for visual recognition while managing computational cost.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning Algorithms in Depth. The section links point to the parts of the book it draws on.