DeepPR: Incremental Recovery for Interdependent VNFs with Deep Reinforcement Learning.
Genya IshigakiSiddartha DevicRiti GourJason P. JuePublished in: GLOBECOM (2019)
Keyphrases
- reinforcement learning
- incremental learning
- function approximation
- model free
- learning algorithm
- dynamic programming
- state space
- optimal policy
- temporal difference
- reinforcement learning algorithms
- deep learning
- real world
- recovery algorithm
- multi agent reinforcement learning
- image recovery
- data sets
- error detection
- markov decision process
- temporal difference learning
- learning process
- multi agent
- incremental version
- direct policy search