Adversarial Reinforcement Learning under Partial Observability in Autonomous Computer Network Defence.
Yi HanDavid HubczenkoPaul MontagueOlivier Y. de VelTamas AbrahamBenjamin I. P. RubinsteinChristopher LeckieTansu AlpcanSarah M. ErfaniPublished in: IJCNN (2020)
Keyphrases
- partial observability
- computer networks
- reinforcement learning
- partially observable
- multi agent
- markov decision process
- belief state
- intrusion detection system
- planning problems
- belief space
- state space
- network traffic
- markov decision processes
- computer systems
- partially observable markov decision processes
- model free
- machine learning
- function approximation
- dynamic programming
- learning agent
- reinforcement learning algorithms
- learning capabilities
- computer network security
- mobile robot
- partial information
- intrusion detection
- temporal difference
- optimal policy
- dynamical systems