Adversarial Manipulation of Reinforcement Learning Policies in Autonomous Agents.
Yonghong HuangShih-Han WangPublished in: IJCNN (2018)
Keyphrases
- autonomous agents
- reinforcement learning
- multi agent
- optimal policy
- multi agent environments
- policy search
- learning agents
- markov decision process
- reward function
- multi agent systems
- multiagent systems
- partially observable markov decision processes
- state space
- incomplete information
- control policy
- dynamic environments
- resource bounded
- fitted q iteration
- markov decision processes
- multiple agents
- function approximation
- dynamic programming
- changing environment
- markov decision problems
- complex environments
- action selection
- intelligent behavior
- bdi agents
- open systems
- heterogeneous agents
- reinforcement learning algorithms
- distributed artificial intelligence
- cooperative
- temporal difference
- control architecture
- coalition formation
- intelligent agents
- learning algorithm
- reinforcement learning methods
- single agent
- computational systems
- machine learning
- model free
- coordination mechanisms