Tactical Reward Shaping: Bypassing Reinforcement Learning with Strategy-Based Goals.
Yizheng ZhangAndre RosendoPublished in: CoRR (2019)
Keyphrases
- reward shaping
- reinforcement learning
- reinforcement learning algorithms
- complex domains
- state space
- strategic and tactical
- learning algorithm
- markov decision problems
- function approximation
- neural network
- markov decision processes
- temporal difference
- supervised learning
- transition model
- optimal control
- transfer learning
- action space
- case based reasoning
- dynamic programming