Sandboxing (AI-based) Unverified Controllers in Stochastic Games: An Abstraction-based Approach with Safe-visor Architecture.
Bingzhuo ZhongHongpeng CaoMajid ZamaniMarco CaccamoPublished in: CoRR (2022)
Keyphrases
- stochastic games
- artificial intelligence
- nash equilibria
- expert systems
- multiagent reinforcement learning
- multi agent
- markov decision processes
- average reward
- reinforcement learning algorithms
- machine learning
- knowledge based systems
- repeated games
- case based reasoning
- control system
- reinforcement learning
- information processing
- nash equilibrium
- dynamic programming
- imperfect information
- cooperative
- learning algorithm