{
  "scenario_id": "routing_modelos_soporte_c03",
  "status": "pass",
  "selected_policy": "greedy",
  "rounds": 60,
  "contract_version": "bandit_policy_gate.v1",
  "policies": [
    {
      "policy_id": "greedy",
      "cumulative_reward": 46.63,
      "regret": 0.27,
      "exploration_share": 0.05,
      "sensitive_exploration_count": 0,
      "average_cost": 0.1865,
      "action_counts": {
        "modelo_fuerte": 58,
        "modelo_rapido": 1,
        "revision_humana": 1
      },
      "action_share": {
        "modelo_fuerte": 0.9667,
        "modelo_rapido": 0.0167,
        "revision_humana": 0.0167
      },
      "observed_means": {
        "modelo_fuerte": 0.7824,
        "modelo_rapido": 0.6,
        "revision_humana": 0.65
      },
      "gate_checks": {
        "min_cumulative_reward": true,
        "max_regret": true,
        "max_exploration_share": true,
        "max_sensitive_exploration_count": true,
        "max_average_cost": true,
        "required_trace_events": true
      },
      "gate_ok": true
    },
    {
      "policy_id": "epsilon_greedy",
      "cumulative_reward": 45.95,
      "regret": 0.95,
      "exploration_share": 0.1333,
      "sensitive_exploration_count": 0,
      "average_cost": 0.1973,
      "action_counts": {
        "modelo_fuerte": 53,
        "modelo_rapido": 4,
        "revision_humana": 3
      },
      "action_share": {
        "modelo_fuerte": 0.8833,
        "modelo_rapido": 0.0667,
        "revision_humana": 0.05
      },
      "observed_means": {
        "modelo_fuerte": 0.7836,
        "modelo_rapido": 0.6025,
        "revision_humana": 0.67
      },
      "gate_checks": {
        "min_cumulative_reward": true,
        "max_regret": true,
        "max_exploration_share": true,
        "max_sensitive_exploration_count": true,
        "max_average_cost": true,
        "required_trace_events": true
      },
      "gate_ok": true
    },
    {
      "policy_id": "ucb",
      "cumulative_reward": 43.08,
      "regret": 3.82,
      "exploration_share": 0.4833,
      "sensitive_exploration_count": 0,
      "average_cost": 0.2927,
      "action_counts": {
        "modelo_fuerte": 32,
        "modelo_rapido": 12,
        "revision_humana": 16
      },
      "action_share": {
        "modelo_fuerte": 0.5333,
        "modelo_rapido": 0.2,
        "revision_humana": 0.2667
      },
      "observed_means": {
        "modelo_fuerte": 0.7897,
        "modelo_rapido": 0.5925,
        "revision_humana": 0.6687
      },
      "gate_checks": {
        "min_cumulative_reward": true,
        "max_regret": true,
        "max_exploration_share": false,
        "max_sensitive_exploration_count": true,
        "max_average_cost": false,
        "required_trace_events": true
      },
      "gate_ok": false
    },
    {
      "policy_id": "thompson_sampling",
      "cumulative_reward": 46.63,
      "regret": 0.27,
      "exploration_share": 0.05,
      "sensitive_exploration_count": 0,
      "average_cost": 0.1865,
      "action_counts": {
        "modelo_fuerte": 58,
        "modelo_rapido": 1,
        "revision_humana": 1
      },
      "action_share": {
        "modelo_fuerte": 0.9667,
        "modelo_rapido": 0.0167,
        "revision_humana": 0.0167
      },
      "observed_means": {
        "modelo_fuerte": 0.7824,
        "modelo_rapido": 0.6,
        "revision_humana": 0.65
      },
      "gate_checks": {
        "min_cumulative_reward": true,
        "max_regret": true,
        "max_exploration_share": true,
        "max_sensitive_exploration_count": true,
        "max_average_cost": true,
        "required_trace_events": true
      },
      "gate_ok": true
    }
  ]
}