Offline policy evaluation across representations with applications to educational games.
Travis MandelYun-En LiuSergey LevineEmma BrunskillZoran PopovicPublished in: AAMAS (2014)
Keyphrases
- educational games
- policy evaluation
- least squares
- video games
- temporal difference
- learning tools
- model free
- monte carlo
- markov decision processes
- reinforcement learning
- computer games
- policy iteration
- semi parametric
- function approximation
- game development
- optimal policy
- dynamic programming
- partially observable markov decision processes
- search space