lunar lander tuned, 1e6 timesteps, params: {'n_steps': 1024, 'n_epochs': 20, 'discount_factor_gamma': 0.999}

lunar lander tuned, 1e6 timesteps, params: {'n_steps': 4096, 'n_epochs': 20, 'discount_factor_gamma': 0.999}

lunar lander tuned, 1e6 timesteps, params: {'n_steps': 1024, 'n_epochs': 5, 'discount_factor_gamma': 0.999}

lunar lander default training, 1e6 timesteps