{ "baseline": { "mean_reward": -35.27595965255516, "std_reward": 307.12488362532207, "positive_pct": 67.2, "mean_latency_delta": 11.184, "action_diversity": 1 }, "heuristic": { "mean_reward": -35.29905965255515, "std_reward": 307.1265379670056, "positive_pct": 67.2, "mean_latency_delta": 11.184, "action_diversity": 2 }, "ai_strategist": { "mean_reward": -35.276259652555154, "std_reward": 307.12486193924184, "positive_pct": 67.2, "mean_latency_delta": 11.184, "action_diversity": 1 } }