{ "mean_reward": -26.89202117919922, "eval_mean_best": -26.954320907592773, "is_deterministic": true, "D": 64, "seed": 7, "algo": "Hybrid-HD-PPO", "env": "Pusher-v5", "beta_base": 0.7071067690849304, "beta_bands": null, "std_reward": 3.6587091856335845, "ci95_reward": 1.6035, "n_eval_episodes": 20 }