{
  "task": "PPO CartPole clip ablation and SAC PointControl temperature ablation",
  "rerun": "3-seed 12k runs and a seed-11 100k capability run were regenerated from the published configuration",
  "selection_rule": "comparison arms use fixed seed 1 and environment seed 777; the 100k capability run uses predeclared seed 11",
  "files": [
    {
      "file": "ppo_clip0.05_seed1.mp4",
      "type": "video/mp4",
      "method": "PPO clip=.05",
      "seed": 1,
      "metric": "3-seed final return 46.42±8.35",
      "bytes": 35091
    },
    {
      "file": "ppo_clip0.2_seed1.mp4",
      "type": "video/mp4",
      "method": "PPO clip=.2",
      "seed": 1,
      "metric": "3-seed final return 68.18±18.63",
      "bytes": 38498
    },
    {
      "file": "ppo_clip0.4_seed1.mp4",
      "type": "video/mp4",
      "method": "PPO clip=.4",
      "seed": 1,
      "metric": "3-seed final return 68.90±3.91",
      "bytes": 24613
    },
    {
      "file": "sac_fixed_seed1.mp4",
      "type": "video/mp4",
      "method": "SAC fixed alpha=.2",
      "seed": 1,
      "metric": "3-seed final return 166.86±5.62",
      "bytes": 10970
    },
    {
      "file": "sac_auto_seed1.mp4",
      "type": "video/mp4",
      "method": "SAC auto alpha",
      "seed": 1,
      "metric": "3-seed final return 154.00±21.08",
      "bytes": 11127
    },
    {
      "file": "ppo_cartpole_solved_seed11.mp4",
      "type": "video/mp4",
      "method": "PPO capability run",
      "seed": 11,
      "metric": "best rolling return 396.8; last 263.35",
      "bytes": 22062
    }
  ]
}