IMA-GNN: In-Memory Acceleration of Centralized and Decentralized Graph Neural Networks at the Edge.
Mehrdad MorsaliMahmoud NazzalAbdallah KhreishahShaahin AngiziPublished in: CoRR (2023)
Keyphrases
- neural network
- weighted graph
- peer to peer
- auto associative
- edge weights
- peer to peer systems
- undirected graph
- disjoint paths
- structured data
- associative memory
- distributed systems
- pattern recognition
- fuzzy logic
- multi agent
- graph representation
- bipartite graph
- vertex set
- cooperative
- graph model
- graph structure
- random walk
- artificial neural networks
- memory requirements
- edge detection
- directed graph
- distributed environment
- neural nets
- genetic algorithm
- strongly connected
- self organizing maps
- graph theoretic
- spanning tree
- directed acyclic graph
- neural network model
- edge information
- digital libraries
- training process
- memory usage
- main memory
- fault diagnosis
- data objects
- fully distributed
- graph theory
- graph based algorithm
- back propagation