Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning.
Yinglun XuGagandeep SinghPublished in: CoRR (2023)
Keyphrases
- black box
- reinforcement learning
- black boxes
- white box
- online learning
- state space
- function approximation
- state transition
- model free
- hybrid systems
- reinforcement learning algorithms
- white box testing
- average reward
- neural network
- reward function
- machine learning
- database
- action selection
- temporal difference
- learning algorithm
- integration testing
- eligibility traces
- case study
- markov decision process
- decision trees