RLDS: an Ecosystem to Generate, Share and Use Datasets in Reinforcement Learning.
Sabela RamosSertan GirginLéonard HussenotDamien VincentHanna YakubovichDaniel ToyamaAnita GergelyPiotr StanczykRaphaël MarinierJeremiah HarmsenOlivier PietquinNikola MomchevPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- real life
- dynamic programming
- automatically generate
- model free
- benchmark datasets
- markov decision processes
- reinforcement learning algorithms
- uci machine learning repository
- neural network
- machine learning
- genetic algorithm
- learning process
- function approximation
- training dataset
- synthetic and real datasets