Real-Time Adversarial Perturbations Against Deep Reinforcement Learning Policies: Attacks and Defenses.
Buse G. A. TekgulShelly WangSamuel MarchalN. AsokanPublished in: ESORICS (3) (2022)
Keyphrases
- real time
- reinforcement learning
- optimal policy
- denial of service attacks
- multi agent
- robotic control
- low cost
- high speed
- policy search
- markov decision processes
- countermeasures
- function approximation
- markov decision process
- model free
- network security
- partially observable markov decision processes
- control policies
- quality of service
- dos attacks
- continuous state
- computer networks
- neural network