SESKStuff Everyone Should Know
SESK: Stuff Everyone Should Know · sesk.info/ai-cnn-shared-filter-weights

How does a convolutional layer reduce parameters across image locations?

It reuses the same filter weights at many locations.

Start playing free

  1. By sharing filter weights ✔
  2. By deleting image channels
  3. By freezing every bias

Why: One learned filter slides across the image instead of learning a separate detector at every position. This greatly reduces the parameter count.

Huh, didn't know that: Shared filters can detect the same kind of edge in different parts of an image.

The dad joke
The CNN shared its weights, but the gym still charged full price.
Like this card?
Play the game

References

Answer: By sharing filter weights. One learned filter slides across the image instead of learning a separate detector at every position. This greatly reduces the parameter count.

  1. Stanford CS231n: Convolutional Networks cs231n.github.io/convolutional-networks
    constrain the neurons in each depth slice to use the same weights and bias
    Checked 2026-10-10.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Test yourself in the game

Artificial IntelligenceIn a CNN, what does a neuron's receptive field describe?Artificial IntelligenceHow do AI, machine learning and deep learning fit together?Artificial IntelligenceWhat did Alan Turing call his 1950 test of machine intelligence?

That's stuff everyone should know.