Grounding Natural Language Commands to StarCraft II Game States for Narration-Guided Reinforcement Learning.
Nicholas R. WaytowichSean L. BartonVernon LawhernEthan StumpGarrett WarnellPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- natural language
- perceptual aliasing
- video games
- game play
- opponent modeling
- game playing
- interactive storytelling
- computer games
- machine learning
- game theory
- nash equilibrium
- state abstraction
- state action
- function approximation
- temporal difference learning
- language processing
- semantic analysis
- educational games
- transition model
- state space
- reinforcement learning algorithms
- stochastic games
- markov decision problems
- semantic interpretation
- dynamic programming
- learning algorithm
- knowledge representation
- natural language processing
- model free
- game design
- learning process
- serious games
- natural language interface
- optimal policy
- hidden markov models
- information extraction
- virtual world
- learning experience
- markov decision processes
- nash equilibria
- initial state
- natural language generation
- digital games
- state transition