ScholarGate
Assistent
Machine learning

Capsule Network

Et Capsule Network (CapsNet) er en deep learning-arkitektur introduceret af Sara Sabour, Nicholas Frosst og Geoffrey Hinton i 2017, som organiserer neuroner som vektorer (kapsler) i stedet for skalaraktiveringer, således at rumlig hierarki og pose (orientering) information kodes direkte. Den blev foreslået for at overvinde den skrøbelighed, som konvolutionelle netværk udviser over for ændringer i synsvinkel.

Åbn i MethodMindSnartVideoSnartDownload slides

Læs hele metoden

Kun for medlemmer

Log ind med en gratis konto for at læse dette afsnit.

Log ind

Method map

The neighbourhood of related methods — select a node to explore.

Kilder

  1. Sabour, S., Frosst, N. & Hinton, G. E. (2017). Dynamic Routing Between Capsules. Advances in Neural Information Processing Systems (NeurIPS). link
  2. Hinton, G. E., Sabour, S. & Frosst, N. (2018). Matrix Capsules with EM Routing. International Conference on Learning Representations (ICLR). link

Sådan citerer du denne side

ScholarGate. (2026, June 1). Capsule Network (CapsNet). ScholarGate. https://scholargate.app/da/deep-learning/capsule-network

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

Compare side by side

Refereret af

ScholarGateCapsule Network (Capsule Network (CapsNet)). Hentet 2026-06-15 fra https://scholargate.app/da/deep-learning/capsule-network · Datasæt: https://doi.org/10.5281/zenodo.20539026