Google research football : a novel reinforcement learning environment
| Autor: | Damien Vincent, Olivier Bachem, Karol Kurach, Piotr Stańczyk, Anton Raichuk, Michał Zając, Marcin Michalski, Sylvain Gelly, Lasse Espeholt, Olivier Bousquet, Carlos Riquelme |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning 0209 industrial biotechnology Computer science ComputingMilieux_PERSONALCOMPUTING Machine Learning (stat.ML) 02 engineering and technology General Medicine Football Field (computer science) Machine Learning (cs.LG) 020901 industrial engineering & automation Statistics - Machine Learning Human–computer interaction 0202 electrical engineering electronic engineering information engineering Virtual learning environment Reinforcement learning 020201 artificial intelligence & image processing Set (psychology) Baseline (configuration management) Reinforcement |
| Zdroj: | AAAI |
| Popis: | Recent progress in the field of reinforcement learning has been accelerated by virtual learning environments such as video games, where novel algorithms and ideas can be quickly tested in a safe and reproducible manner. We introduce the Google Research Football Environment, a new reinforcement learning environment where agents are trained to play football in an advanced, physics-based 3D simulator. The resulting environment is challenging, easy to use and customize, and it is available under a permissive open-source license. In addition, it provides support for multiplayer and multi-agent experiments. We propose three full-game scenarios of varying difficulty with the Football Benchmarks and report baseline results for three commonly used reinforcement algorithms (IMPALA, PPO, and Ape-X DQN). We also provide a diverse set of simpler scenarios with the Football Academy and showcase several promising research directions. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|