A Minimal Approach for Natural Language Action Space in Text-based Games.
Dongwon Kelvin RyuMeng FangShirui PanGholamreza HaffariEhsan ShareghiPublished in: CoRR (2023)
Keyphrases
- action space
- natural language
- state space
- markov decision processes
- real valued
- reinforcement learning
- state and action spaces
- stochastic games
- stochastic processes
- action selection
- nash equilibria
- machine learning
- game theory
- state action
- single agent
- markov chain
- np hard
- finite state
- hill climbing
- markov decision process
- hidden markov models
- continuous state spaces