State-Space Closure: Revisiting Endless Online Level Generation via Reinforcement Learning.
Ziqi WangTianye ShuJialin LiuPublished in: IEEE Trans. Games (2024)
Keyphrases
- state space
- reinforcement learning
- reinforcement learning algorithms
- markov decision processes
- function approximation
- optimal policy
- online learning
- markov chain
- dynamic programming
- state variables
- heuristic search
- learning algorithm
- action space
- temporal difference
- dynamical systems
- planning problems
- belief state
- model free
- markov decision process
- real time
- partially observable
- levels of abstraction
- finite state
- state and action spaces
- balancing exploration and exploitation
- generation process
- goal state
- temporal difference learning
- continuous state spaces
- stochastic domains
- learning agent
- initial state
- control problems
- multi agent
- learning process
- supervised learning
- particle filter
- fixed point
- higher level