Index Ai Academy
Glossary · Foundation · Foundation math

Weight sharing

Also written: shared weights, parameter sharing

Using the same few weights at every place in the input, as a convolution does when it slides one pattern everywhere. It needs far fewer weights than a plain layer, and a pattern learned in one place works in all of them.

Try it now

Then set stride s to 2 and press Use these numbers. The pattern jumps two places at a time, and only 2 stops fit.

Learn it properly

Sliding dot products: convolution · A small pattern slid along a list and over a grid, output sizes, stride and padding, and shared weights. About 40 minutes.

Other terms from the same lesson

  • convolution: Sliding a small pattern of weights along a list or across an image and, at every stop, multiplying each weight by the value under it and adding up. One pattern finds the same feature, such as an edge, wherever it is.
Keep learning

More in plain English.

1286 terms in the glossary, each with the lesson that teaches it.

1286 terms

The glossary

Every word you'll meet in AI and machine learning, explained in everyday language.

Foundation · lesson 40 min

Sliding dot products: convolution

A small pattern slid along a list and over a grid, output sizes, stride and padding, and shared weights.

Never trained an AI? Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account. Teach it to write