On the use of graph neural networks and shape-function-based gradient computation in the deep energy method.
Junyan HeDiab W. AbueiddaSeid KoricIwona JasiukPublished in: CoRR (2022)
Keyphrases
- neural network
- cost function
- pattern recognition
- fuzzy logic
- graph construction
- probabilistic model
- statistical shape model
- edge detection
- shortest path
- signed distance function
- graph representation
- training process
- density function
- graph theory
- prior information
- energy function
- random walk
- support vector machine
- artificial neural networks
- clustering algorithm