Characterizing Attacks on Deep Reinforcement Learning.
Chaowei XiaoXinlei PanWarren HeJian PengMingjie SunJinfeng YiMingyan LiuBo LiDawn SongPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- countermeasures
- markov decision processes
- state space
- model free
- traffic analysis
- multi agent
- reinforcement learning algorithms
- watermarking scheme
- optimal control
- optimal policy
- transfer learning
- malicious attacks
- machine learning
- deep learning
- temporal difference learning
- policy search
- temporal difference
- malicious users
- learning classifier systems
- learning problems
- data sets
- watermarking algorithm
- security mechanisms
- supervised learning
- cryptographic protocols
- multi agent systems
- information systems
- denial of service attacks
- learning algorithm
- transition model
- genetic algorithm
- robotic control
- detecting malicious