Develops an unsupervised training algorithm for capsule networks by aligning an energy function with dynamic routing to enable log-likelihood optimization and image generation on vision datasets.
Capsulegan: Generative adversarial capsule network
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2019 2verdicts
UNVERDICTED 2representative citing papers
Capsule networks with dynamic routing are shown to be equivalent to a routing-weighted product of expert neurons, supporting a bottom-up unsupervised learning algorithm via alternating routing and contrastive divergence.
citing papers explorer
-
Training products of expert capsules with mixing by dynamic routing
Develops an unsupervised training algorithm for capsule networks by aligning an energy function with dynamic routing to enable log-likelihood optimization and image generation on vision datasets.
-
Training capsules as a routing-weighted product of expert neurons
Capsule networks with dynamic routing are shown to be equivalent to a routing-weighted product of expert neurons, supporting a bottom-up unsupervised learning algorithm via alternating routing and contrastive divergence.