Refine
Year of publication
Document Type
- Conference Proceeding (32)
- Article (unreviewed) (1)
- Report (1)
Conference Type
- Sonstiges (18)
- Konferenz-Abstract (12)
- Konferenzartikel (2)
Has Fulltext
- no (34)
Is part of the Bibliography
- yes (34) (remove)
Keywords
- RoboCup (30)
- Deep Reinforcement Learning (1)
Institute
- Fakultät Elektrotechnik und Informationstechnik (E+I) (bis 03/2019) (20)
- Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019) (14)
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (9)
- Fakultät Medien und Informationswesen (M+I) (bis 21.04.2021) (2)
- IMLA - Institute for Machine Learning and Analytics (2)
- Fakultät Wirtschaft (W) (1)
Open Access
- Bronze (34) (remove)
This paper presents the new Deep Reinforcement Learning (DRL) library RL-X and its application to the RoboCup Soccer Simulation 3D League and classic DRL benchmarks. RL-X provides a flexible and easy-to-extend codebase with self-contained single directory algorithms. Through the fast JAX-based implementations, RL-X can reach up to 4.5x speedups compared to well-known frameworks like Stable-Baselines3.