Reward Shaping for Reinforcement Learning with Omega-Regular Objectives.
Ernst Moritz HahnMateo PerezSven ScheweFabio SomenziAshutosh TrivediDominik WojtczakPublished in: CoRR (2020)
Keyphrases
- reward shaping
- reinforcement learning
- reinforcement learning algorithms
- complex domains
- function approximation
- markov decision problems
- state space
- temporal difference
- learning algorithm
- partially observable
- markov decision processes
- neural network
- multiple objectives
- multi agent
- machine learning
- model free
- transfer learning
- reward function
- optimal policy
- markov decision process
- supervised learning
- dynamic programming