Defense Against Reward Poisoning Attacks in Reinforcement Learning.
Kiarash BanihashemAdish SinglaGoran RadanovicPublished in: CoRR (2021)
Keyphrases
- long run
- reinforcement learning
- optimal policy
- average reward
- defense mechanisms
- ddos attacks
- markov decision processes
- state space
- function approximation
- countermeasures
- machine learning
- reinforcement learning algorithms
- network security
- model free
- reward function
- malicious attacks
- denial of service attacks
- temporal difference
- learning algorithm
- dos attacks
- computer virus
- state action
- security mechanisms
- function approximators
- advanced research projects agency
- policy gradient
- security protocols
- computer security
- learning process
- intrusion detection
- learning agent
- multi agent
- learning capabilities
- security threats
- action selection
- dynamic programming
- eligibility traces
- partially observable environments
- data mining
- optimal control