Accelerating Large Sparse Neural Network Inference Using GPU Task Graph Parallelism.
Dian-Lun LinTsung-Wei HuangPublished in: IEEE Trans. Parallel Distributed Syst. (2022)
Keyphrases
- neural network
- parallel processing
- parallel computing
- parallel computation
- gaussian graphical models
- graph representation
- artificial neural networks
- real time
- directed acyclic
- sparse data
- bayesian networks
- graph model
- graph theory
- probabilistic inference
- fuzzy logic
- gpu implementation
- spanning tree
- parallel implementation
- belief networks
- high dimensional
- directed graph
- recurrent neural networks
- general purpose
- neural network is trained
- clique tree
- back propagation
- parallel execution
- sparse representation
- connected components
- bayesian inference
- bipartite graph
- computational power
- conditional random fields
- radial basis function
- graphics processing units
- genetic algorithm
- parallel programming
- neural network model
- multithreading
- directed acyclic graph
- weighted graph
- signal recovery
- random walk