{"event_id":"bad_001","context":{"slice":"baja_criticidad","complexity":0.20},"allowed_actions":["modelo_rapido","modelo_fuerte","revision_humana"],"action":"modelo_rapido","behavior_action_probability":0.90,"target_action_probability":0.05,"target_policy_probability_by_action":{"modelo_rapido":0.05,"modelo_fuerte":0.90,"revision_humana":0.05},"q_model_reward_by_action":{"modelo_rapido":0.70,"modelo_fuerte":0.72,"revision_humana":0.58},"reward":0.69}
{"event_id":"bad_002","context":{"slice":"baja_criticidad","complexity":0.30},"allowed_actions":["modelo_rapido","modelo_fuerte","revision_humana"],"action":"modelo_rapido","behavior_action_probability":0.88,"target_action_probability":0.04,"target_policy_probability_by_action":{"modelo_rapido":0.04,"modelo_fuerte":0.91,"revision_humana":0.05},"q_model_reward_by_action":{"modelo_rapido":0.67,"modelo_fuerte":0.74,"revision_humana":0.60},"reward":0.68}
{"event_id":"bad_003","context":{"slice":"media_criticidad","complexity":0.60},"allowed_actions":["modelo_rapido","modelo_fuerte","revision_humana"],"action":"modelo_rapido","behavior_action_probability":0.82,"target_action_probability":0.03,"target_policy_probability_by_action":{"modelo_rapido":0.03,"modelo_fuerte":0.92,"revision_humana":0.05},"q_model_reward_by_action":{"modelo_rapido":0.58,"modelo_fuerte":0.78,"revision_humana":0.70},"reward":0.55}
{"event_id":"bad_004","context":{"slice":"media_criticidad","complexity":0.64},"allowed_actions":["modelo_rapido","modelo_fuerte","revision_humana"],"action":"revision_humana","behavior_action_probability":0.03,"target_action_probability":0.95,"target_policy_probability_by_action":{"modelo_rapido":0.02,"modelo_fuerte":0.03,"revision_humana":0.95},"q_model_reward_by_action":{"modelo_rapido":0.56,"modelo_fuerte":0.78,"revision_humana":0.71},"reward":0.72}
{"event_id":"bad_005","context":{"slice":"alta_criticidad","complexity":0.90},"allowed_actions":["modelo_fuerte","revision_humana"],"action":"revision_humana","behavior_action_probability":0.04,"target_action_probability":0.96,"target_policy_probability_by_action":{"modelo_fuerte":0.04,"revision_humana":0.96},"q_model_reward_by_action":{"modelo_fuerte":0.80,"revision_humana":0.76},"reward":0.78}
{"event_id":"bad_006","context":{"slice":"alta_criticidad","complexity":0.88},"allowed_actions":["modelo_fuerte","revision_humana"],"action":"revision_humana","behavior_action_probability":0.05,"target_action_probability":0.95,"target_policy_probability_by_action":{"modelo_fuerte":0.05,"revision_humana":0.95},"q_model_reward_by_action":{"modelo_fuerte":0.83,"revision_humana":0.75},"reward":0.76}
