BadRL: Sparse Targeted Backdoor Attack Against Reinforcement Learning.
Jing CuiYufei HanYuzhe MaJianbin JiaoJunge ZhangPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- high dimensional
- learning algorithm
- reinforcement learning algorithms
- sparse representation
- multi agent
- temporal difference
- countermeasures
- function approximation
- machine learning
- multi agent reinforcement learning
- sparse data
- model free
- markov decision processes
- dynamic programming
- optimal policy
- state space
- temporal difference learning
- neural network
- sparse coding
- optimal control
- supervised learning
- mobile robot
- learning process
- compressed sensing
- function approximators
- robotic control