Minimalistic Attacks: How Little It Takes to Fool Deep Reinforcement Learning Policies.
Xinghua QuZhu SunYew-Soon OngAbhishek GuptaPengfei WeiPublished in: IEEE Trans. Cogn. Dev. Syst. (2021)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- control policies
- state space
- reinforcement learning agents
- reward function
- partially observable markov decision processes
- hierarchical reinforcement learning
- countermeasures
- policy gradient methods
- markov decision processes
- markov decision problems
- model free
- reinforcement learning algorithms
- control policy
- fitted q iteration
- function approximation
- learning algorithm
- learning classifier systems
- watermarking scheme
- machine learning
- traffic analysis
- terrorist attacks
- multi agent
- watermarking algorithm
- security threats
- denial of service attacks
- dynamic programming
- neural network