Scalable Heterogeneous Scheduling Based Model Parallelism for Real-Time Inference of Large-Scale Deep Neural Networks.
Xiaofeng ZouCen ChenPeiying LinLuochuan ZhangYanwu XuWenjie ZhangPublished in: IEEE Trans. Emerg. Top. Comput. Intell. (2024)
Keyphrases
- real time
- neural network
- high level
- computational model
- real world
- neural network model
- resource allocation
- statistical model
- management system
- experimental data
- probability distribution
- probabilistic model
- artificial neural networks
- mathematical model
- pattern recognition
- similarity measure
- prediction model
- multi layer
- web scale