Dynamic Semantic Compression for CNN Inference in Multi-access Edge Computing: A Graph Reinforcement Learning-based Autoencoder.
Nan LiAlexandros IosifidisQi ZhangPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- weighted graph
- random access
- undirected graph
- graph model
- directed graph
- data compression
- hyperspectral image compression
- dynamic graph
- high level
- compression ratio
- natural language
- edge detection
- function approximation
- edge weights
- graph representation
- edge information
- graph structure
- vertex set
- structured data
- dynamic environments
- random walk
- access control
- homology generators
- probabilistic inference
- cellular neural networks
- neural network
- data structure
- bayesian networks
- machine learning