Capsule network with shortcut routing
This addresses computational bottlenecks in capsule networks for hierarchical pattern representation, though it appears incremental as it builds on existing routing mechanisms.
The study tackled computational inefficiency in capsule networks by introducing shortcut routing, which directly activates global capsules from local capsules to eliminate intermediate layers, achieving comparable classification performance (99.52% on Mnist, 93.91% on smallnorb, 89.02% on affNist) while reducing calculations by 1.42-2.5 times compared to EM routing.
This study introduces "shortcut routing," a novel routing mechanism in capsule networks that addresses computational inefficiencies by directly activating global capsules from local capsules, eliminating intermediate layers. An attention-based approach with fuzzy coefficients is also explored for improved efficiency. Experimental results on Mnist, smallnorb, and affNist datasets show comparable classification performance, achieving accuracies of 99.52%, 93.91%, and 89.02% respectively. The proposed fuzzy-based and attention-based routing methods significantly reduce the number of calculations by 1.42 and 2.5 times compared to EM routing, highlighting their computational advantages in capsule networks. These findings contribute to the advancement of efficient and accurate hierarchical pattern representation models.