DARe: DropLayer-Aware Manycore ReRAM architecture for Training Graph Neural Networks.
Aqeeb Iqbal ArkaBiresh Kumar JoardarJanardhan Rao DoppaPartha Pratim PandeKrishnendu ChakrabartyPublished in: ICCAD (2021)
Keyphrases
- neural network
- training process
- training algorithm
- neural network structure
- network architecture
- multi layer
- feedforward neural networks
- pattern recognition
- multi layer perceptron
- back propagation
- random walk
- backpropagation algorithm
- error back propagation
- feed forward neural networks
- neural network training
- real time
- structured data
- graph theoretic
- training phase
- learning capabilities
- graph model
- feedforward artificial neural networks
- recurrent networks
- multilayer neural network
- weighted graph
- fuzzy logic
- fault diagnosis
- artificial neural networks
- associative memory
- graph theory
- test set
- software architecture
- multilayer perceptron
- recurrent neural networks
- graph partitioning
- directed graph
- graph structure
- directed acyclic graph
- radial basis function network
- bipartite graph
- decision trees
- training data
- genetic algorithm
- graphical models
- parallel algorithm
- connected components
- graph representation
- neural nets