Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations.
Anuroop SriramAbhishek DasBrandon M. WoodSiddharth GoyalC. Lawrence ZitnickPublished in: CoRR (2022)
Keyphrases
- neural network
- training process
- training algorithm
- feed forward neural networks
- feedforward neural networks
- multi layer perceptron
- pattern recognition
- back propagation
- graph theory
- graph representation
- backpropagation algorithm
- graph theoretic
- radial basis function network
- recurrent networks
- neural network structure
- genetic algorithm
- error back propagation
- multi layer
- fuzzy logic
- multilayer perceptron
- random walk
- directed graph
- training set
- neural network training
- connected components
- neural network model
- directed acyclic graph
- parameter values
- neural nets
- parameter space
- bipartite graph
- feed forward
- training examples
- training samples
- supervised learning
- graph structure