BitGNN: Unleashing the Performance Potential of Binary Graph Neural Networks on GPUs.
Jou-An ChenHsin-Hsuan SungXipeng ShenSutanay ChoudhuryAng LiPublished in: ICS (2023)
Keyphrases
- neural network
- data sets
- pattern recognition
- artificial neural networks
- graph theory
- connected components
- graph theoretic
- graph representation
- graph model
- graph structure
- real time
- random walk
- structured data
- neural nets
- general purpose
- fuzzy logic
- weighted graph
- genetic algorithm
- dependency graph
- fault diagnosis
- neural network model
- parallel processing
- training process
- directed acyclic graph
- pairwise
- learning algorithm
- graphics hardware