Grounding English Commands to Reward Functions.
James MacGlashanMonica Babes-VromanMarie desJardinsMichael L. LittmanSmaranda MuresanShawn SquireStefanie TellexDilip ArumugamLei YangPublished in: Robotics: Science and Systems (2015)
Keyphrases
- reward function
- reinforcement learning
- state space
- markov decision processes
- multiple agents
- inverse reinforcement learning
- machine translation
- transition probabilities
- natural language
- optimal policy
- simple examples
- markov decision process
- state variables
- policy search
- cross lingual
- markov decision problems
- generative model
- transition model
- multi agent