Announcement_arlbench

Our joint work about the Automated RL Benchmark was accepted at the European Workshop of Reinforcement Learning this year and is now available at Arxiv. The benchmark contains lightning fast implementations of three common RL algorithms in JAX, build specifically to benchmark AutoRL methods in a consistent setting.