{
  "env": "cheetah",
  "title": "SHAC with MuJoCo Warp",
  "timestamp_pacific": "2026-06-17T14:31:33-07:00",
  "mujoco_warp_pr": "google-deepmind/mujoco_warp#1423",
  "newton_commit": "11920df6a589df9693046a3401f455f651a06750",
  "newton_path": "/home/horde/repos/newton-shac/newton",
  "mujoco_warp_commit": "11920df6a589df9693046a3401f455f651a06750",
  "num_envs": 1024,
  "selection_num_envs": 1024,
  "seed": 7,
  "contact_backend": "mujoco",
  "horizon": 32,
  "epochs": 0,
  "dt": 0.016666666666666666,
  "sim_substeps": 1,
  "mujoco_integrator": "euler",
  "nconmax": 128,
  "njmax": 512,
  "world_spacing": [
    0.0,
    0.0,
    0.0
  ],
  "force_scale": 200.0,
  "episode_length": 1000,
  "disable_eulerdamp": true,
  "stochastic_init": true,
  "eval_stochastic_init": false,
  "stochastic_actor": true,
  "actor_hidden_dims": null,
  "actor_logstd_init": -1.0,
  "actor_layer_norm": true,
  "critic_hidden_dims": null,
  "use_critic": true,
  "obs_rms": true,
  "actor_path": "shac/assets/cheetah_shac_mujoco_sub1_lr5e4_clip05_term200_h32_select480_16env_60ep_run1/cheetah_actor.pt",
  "obs_rms_path": "shac/assets/cheetah_shac_mujoco_sub1_lr5e4_clip05_term200_h32_select480_16env_60ep_run1/cheetah_obs_rms.pt",
  "critic_method": "td-lambda",
  "td_lambda": 0.95,
  "rew_scale": 1.0,
  "reset_each_epoch": false,
  "termination_penalty": 200.0,
  "terminal_fall_reward": -200.0,
  "selection_fall_penalty": 500000.0,
  "selection_invalid_penalty": 500000.0,
  "selection_displacement_weight": 0.0,
  "selection_height_weight": 0.0,
  "selection_up_weight": 0.0,
  "selection_heading_weight": 0.0,
  "selection_min_height": null,
  "selection_min_up": null,
  "selection_min_heading": null,
  "selection_posture_penalty": 0.0,
  "lr": 0.0005,
  "min_lr": 1e-05,
  "adam_beta1": 0.7,
  "adam_beta2": 0.95,
  "grad_clip": 0.5,
  "grad_value_clip": 1000000.0,
  "critic_lr": 0.002,
  "critic_iterations": 16,
  "critic_batch_size": 1024,
  "target_critic_alpha": 0.2,
  "selection_horizon": 480,
  "ant_asset": null,
  "ant_max_healthy_height": null,
  "ant_termination_height": null,
  "ant_start_height": null,
  "ant_start_joint_q": null,
  "ant_reset_position_scale": null,
  "ant_reset_angle_scale": null,
  "ant_reset_joint_scale": null,
  "ant_reset_velocity_scale": null,
  "ant_density_override": null,
  "ant_height_reward_cap": null,
  "ant_invalid_penalty": null,
  "lr_schedule": "linear",
  "cartpole_reward": {
    "pole_angle": 1.0,
    "pole_velocity": 0.1,
    "cart_position": 0.05,
    "cart_velocity": 0.1,
    "action": 0.0
  },
  "acrobot_reward": {
    "target": 8.0,
    "velocity": 0.05,
    "action": 0.002
  },
  "acrobot_actuation": null,
  "contact_reward": null,
  "ant_reward": {
    "progress": 1.0,
    "heading": 0.5,
    "up": 0.1,
    "height": 1.0,
    "action": 0.0,
    "alive": 0.5,
    "actions_cost": 0.005,
    "energy_cost": 0.05,
    "dof_limit_cost": 1.0,
    "dof_vel_scale": 0.2
  },
  "hopper_reward": null,
  "hopper_terminate_angle": null,
  "hopper_termination_angle": null,
  "hopper_termination_height": null,
  "hopper_termination_height_tolerance": null,
  "hopper_reward_style": null,
  "hopper_start_joint_q": null,
  "hopper_contact_mu": null,
  "hopper_joint_damping": null,
  "hopper_armature": null,
  "hopper_reset_position_scale": null,
  "hopper_reset_angle_scale": null,
  "hopper_reset_joint_scale": null,
  "hopper_reset_velocity_scale": null,
  "cheetah_reward": {
    "action": -0.1
  },
  "ant_disable_joint_limits": null,
  "ant_contact_margin": null,
  "ant_contact_gap": null,
  "ant_contact_mu": null,
  "ant_joint_damping": null,
  "ant_min_up": null,
  "ant_observation_style": null,
  "ant_reward_style": null,
  "ant_action_order": null,
  "phase_observation": null,
  "phase_period": null,
  "locomotion_disable_joint_limits": false,
  "total_seconds": 23.672238574014045,
  "mean_epoch_seconds": null,
  "mean_fps": null,
  "initial_selection": {
    "mean_reward": 1.1434564590454102,
    "return": 548.8590698242188,
    "alive_fraction": 1.0,
    "reset_count": 1024,
    "invalid_count": 0,
    "fall_count": 0,
    "timeout_count": 1024,
    "completed_episodes": 1024,
    "mean_completed_return": 548.8590814471245,
    "mean_completed_length": 480.0,
    "mean_forward_displacement": 10.467256546020508,
    "mean_height": null,
    "mean_up": null,
    "mean_heading": null,
    "unfinished_mean_return": 0.0,
    "unfinished_mean_length": 0.0,
    "final_obs_mean": [
      0.47635096311569214,
      -1.3537843227386475,
      0.17414672672748566,
      -0.07261352986097336,
      0.07264453917741776,
      0.12784729897975922,
      -0.25491029024124146,
      -0.9776685237884521,
      0.5512140989303589,
      -0.041717179119586945,
      -0.09496995061635971,
      -0.03719485551118851,
      0.12074582278728485,
      0.048580098897218704,
      -0.24041013419628143,
      0.11727678775787354,
      0.2732631266117096
    ],
    "final_obs_normalized_mean": [
      -0.21265026926994324,
      -0.9419491291046143,
      -0.2588416337966919,
      -0.14698216319084167,
      0.38371098041534424,
      -0.363447368144989,
      -0.2997358441352844,
      0.14267504215240479,
      0.17458218336105347,
      -0.10103504359722137,
      0.00599020067602396,
      -0.011461596004664898,
      0.014479048550128937,
      0.005391214042901993,
      -0.05089575797319412,
      0.01528172753751278,
      0.03712250292301178
    ]
  },
  "initial_selection_score": 548.8590698242188,
  "best_epoch": 0,
  "best_train_reward": null,
  "best_eval_return": 548.8590698242188,
  "best_eval_score": 548.8590698242188,
  "eval_score": 554.7223510742188,
  "max_cuda_memory_allocated_mb": 11.103515625,
  "max_cuda_memory_reserved_mb": 24.0,
  "history": [],
  "eval": {
    "mean_reward": 1.1556715965270996,
    "return": 554.7223510742188,
    "alive_fraction": 1.0,
    "reset_count": 1024,
    "invalid_count": 0,
    "fall_count": 0,
    "timeout_count": 1024,
    "completed_episodes": 1024,
    "mean_completed_return": 554.722353592515,
    "mean_completed_length": 480.0,
    "mean_forward_displacement": 10.562906265258789,
    "mean_height": null,
    "mean_up": null,
    "mean_heading": null,
    "unfinished_mean_return": 0.0,
    "unfinished_mean_length": 0.0,
    "final_obs_mean": [
      0.4903089106082916,
      -1.2316116094589233,
      0.1795622557401657,
      -0.06264840066432953,
      0.010952351614832878,
      0.1533796787261963,
      -0.21153408288955688,
      -1.038421392440796,
      0.5956159830093384,
      -0.0651267021894455,
      -0.06147538125514984,
      -0.10810539871454239,
      -0.07166449725627899,
      -0.29000020027160645,
      0.05288633704185486,
      -0.17430716753005981,
      0.0011260658502578735
    ],
    "final_obs_normalized_mean": [
      -0.14988306164741516,
      -0.823151707649231,
      -0.243306964635849,
      -0.12670733034610748,
      0.25059813261032104,
      -0.2747395634651184,
      -0.208475261926651,
      0.08032426238059998,
      0.22726187109947205,
      -0.12291662395000458,
      0.022423960268497467,
      -0.02448803372681141,
      -0.009090585634112358,
      -0.027137769386172295,
      0.007316088303923607,
      -0.017442822456359863,
      0.00983384344726801
    ]
  },
  "eval_uninterrupted": {
    "mean_reward": 1.1367536783218384,
    "return": 545.6417236328125,
    "alive_fraction": 1.0,
    "terminal_count": 0,
    "fall_count": 0,
    "invalid_count": 0,
    "first_terminal_step": null,
    "mean_terminal_step": null,
    "mean_forward_displacement": 10.413318634033203,
    "mean_completed_return": 545.6417846679688,
    "mean_height": null,
    "mean_up": null,
    "mean_heading": null,
    "horizon": 480
  },
  "video": "cheetah_rollout.mp4",
  "poster": "cheetah_poster.png",
  "gpu": "NVIDIA RTX PRO 6000 Blackwell Server Edition, [Insufficient Permissions], 580.126.20"
}