Towards verifiable Benchmarks for Reinforcement Learning.
Matthias Müller-BrockhausenAske PlaatMike PreussPublished in: CoG (2022)
Keyphrases
- reinforcement learning
- function approximation
- state space
- optimal policy
- learning algorithm
- sufficient conditions
- model free
- markov decision processes
- relational reinforcement learning
- learning process
- dynamic programming
- learning agents
- transfer learning
- multi agent
- machine learning
- benchmark suite
- direct policy search
- control problems
- real time
- reinforcement learning algorithms
- temporal difference
- information systems
- real world
- data sets