M12.2 CONNECT THE MECHANISM
Slide a shared filter across an input
How does a phone find a letter anywhere in a photo without learning each spot separately? Slide one tiny grid of nine weights across it, and count exactly what comes out.
LESSON OVERVIEW17 min lesson
Lesson overview
How does a phone find a letter anywhere in a photo without learning each spot separately? Slide one tiny grid of nine weights across it, and count exactly what comes out.
What you’ll explore
- Convolutional layers combine local values with shared weights; channels, stride, padding, pooling, and receptive fields determine what information each output can use.
GO TO THE SOURCE
Original explanations, connected to the research.
Dive into Deep Learning — convolutions for imagesDive into Deep Learning — padding and strideDive into Deep Learning — poolingA guide to convolution arithmetic for deep learning (Dumoulin & Visin, 2016)Suggest a correction
A precise note can make an explanation better.
Choose the scene and describe what needs attention. Download a feedback file to share through a channel you already use. This page does not send feedback or connect you with a reviewer.