Reward Delay Attacks on Deep Reinforcement Learning.
Anindya SarkarJiarui FengYevgeniy VorobeychikChristopher D. GillNing ZhangPublished in: GameSec (2022)
Keyphrases
- reinforcement learning
- function approximation
- state space
- reinforcement learning algorithms
- eligibility traces
- countermeasures
- partially observable environments
- markov decision processes
- security threats
- watermarking scheme
- model free
- temporal difference
- learning agent
- optimal policy
- average reward
- total reward
- reward function
- learning algorithm
- machine learning
- partially observable
- transfer learning
- reinforcement learning methods
- malicious users
- security mechanisms
- neural network
- state action
- control system
- traffic analysis
- malicious attacks
- learning process