$Q_{C}-DQN$: A Novel Constrained Reinforcement Learning Method for Computation Offloading in Multi-access Edge Computing.
Shen ZhuangChengxi GaoYing HeF. Richard YuYuhang WangWeike PanZhong MingPublished in: IJCNN (2022)
Keyphrases
- pairwise
- synthetic data
- preprocessing
- high precision
- reinforcement learning
- cost function
- classification method
- objective function
- computational complexity
- detection method
- data sets
- input data
- high accuracy
- experimental evaluation
- optimal policy
- markov random field
- least squares
- computational cost
- mobile robot
- significant improvement
- multi agent
- similarity measure
- image segmentation
- learning algorithm