Rethinking Adversarial Policies: A Generalized Attack Formulation and Provable Defense in Multi-Agent RL.
Xiangyu LiuSouradip ChakrabortyYanchao SunFurong HuangPublished in: CoRR (2023)
Keyphrases
- multi agent
- reinforcement learning
- multiagent reinforcement learning
- optimal policy
- ddos attacks
- defense mechanisms
- cooperative
- intelligent agents
- reinforcement learning agents
- learning agents
- single agent
- multi agent systems
- autonomous agents
- network security
- multiagent systems
- multiple agents
- resolve conflicts
- technology enhanced learning
- partially observable markov decision processes
- markov decision process
- control policies
- intrusion detection
- advanced research projects agency
- policy search
- state space
- reward function
- countermeasures
- continuous state
- function approximation
- markov decision processes