cnn-weight-sharing-parameter-reduction

IN premise — entries/2026/06/21/wiki-Convolutional_neural_network-chunk-1.md

Created 2026-06-21T09:55:49+00:00

CNN weight sharing reduces parameters dramatically: a 5x5 kernel uses 25 weights vs 10,000 for a fully connected layer on 100x100 input

Dependents

These beliefs depend on this one:

• IN cnn-efficiency-from-inductive-biases — CNNs achieve dramatic parameter efficiency through two complementary inductive biases — local connectivity restricts each neuron's receptive field, and weight sharing forces all spatial positions to use identical filters — reducing parameters by orders of magnitude.
• IN weight-sharing-is-universal-efficiency-principle — Weight sharing is a universal efficiency principle appearing independently across architectures — CNN parameter sharing reduces weights by orders of magnitude (25 vs 10,000 for a 5×5 kernel), transformer weight tying halves embedding parameters, and the principle generalizes to any architecture where symmetry in the data structure (spatial, positional) can be exploited to reduce free parameters without reducing expressiveness.