{
  "definition": {
    "T": 100,
    "A_exit": 0.02,
    "C_rates_A_U_L": [
      0.1,
      0.6,
      0.3
    ],
    "U_rates_A_B": [
      1,
      4
    ],
    "L_rates_A_B": [
      7,
      3
    ],
    "B_return": 0.02,
    "cost": "integral u² over active actuator time, arbitrary bookkeeping units",
    "amplitude_cap": 2,
    "delay": "Erlang16 preparation, mean d; canceled on C exit",
    "initial": "A unless specified"
  },
  "baseline": {
    "policy": "无干预",
    "u": 0.0,
    "delay": 0,
    "initial_state": "A",
    "first_B_probability": 0.6723538280186192,
    "B_time": 24.526615230176237,
    "expected_cost": 0.0,
    "final_B_probability": 0.3419987886381991,
    "budget": null,
    "stages": 16,
    "accuracy": 1,
    "false_A": 0
  },
  "static_committor": [
    0.0,
    0.57,
    0.8,
    0.3,
    1.0
  ],
  "finite_terminal_examples": [
    {
      "remaining_time": 0.1,
      "terminal_B_from_C": 0.013067088817046857
    },
    {
      "remaining_time": 1,
      "terminal_B_from_C": 0.3105159310334691
    },
    {
      "remaining_time": 10,
      "terminal_B_from_C": 0.5144753962956917
    },
    {
      "remaining_time": 100,
      "terminal_B_from_C": 0.36696186384867735
    }
  ],
  "expected_budget_examples": [
    {
      "policy": "减少起步",
      "u": 0.11469533323039792,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.6306245040485985,
      "B_time": 22.43854728614748,
      "expected_cost": 0.9999999999999805,
      "final_B_probability": 0.3158368169531022,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 0.7960474742204441,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.5466214850810984,
      "B_time": 18.616673353741895,
      "expected_cost": 0.9999999999998155,
      "final_B_probability": 0.26641030778558605,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "促进返回",
      "u": 0.21128257825898122,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.6723538280186192,
      "B_time": 22.40126915696245,
      "expected_cost": 1.0000000000000098,
      "final_B_probability": 0.3022753356445355,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "减少起步",
      "u": 0.3531077761175254,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.54438199163591,
      "B_time": 18.514055729586737,
      "expected_cost": 10.00000000013951,
      "final_B_probability": 0.26515522185045143,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 2.0,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.4540231564319225,
      "B_time": 14.83581715603145,
      "expected_cost": 6.606620883676392,
      "final_B_probability": 0.21571191458725283,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "促进返回",
      "u": 0.7793368837988413,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.6723538280186192,
      "B_time": 16.464537617882396,
      "expected_cost": 10.000000000000261,
      "final_B_probability": 0.20271367943388346,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "减少起步",
      "u": 0.14872260924416295,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 53.90538046552204,
      "expected_cost": 0.9999999999987392,
      "final_B_probability": 0.35877157120532227,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 1.0361494508187137,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 51.724625822888385,
      "expected_cost": 1.0000000000010203,
      "final_B_probability": 0.31541339554862163,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "促进返回",
      "u": 0.1393310423385423,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 51.511503703672865,
      "expected_cost": 0.9999999999984976,
      "final_B_probability": 0.3476553472075274,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "减少起步",
      "u": 0.4571538840999558,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 51.464129421572466,
      "expected_cost": 10.000000000000075,
      "final_B_probability": 0.3102535410802436,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 2.0,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 50.405695822516364,
      "expected_cost": 3.8279127191569766,
      "final_B_probability": 0.2886024266703181,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "促进返回",
      "u": 0.4872074208626506,
      "delay": 0,
      "initial_state": "B",
      "first_B_probability": 1.0,
      "B_time": 42.12813280149501,
      "expected_cost": 9.999999999946592,
      "final_B_probability": 0.2638932037211781,
      "budget": 10,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    }
  ],
  "expected_budget_atlas": [
    {
      "budget": 0.1,
      "delay": 0,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.25745293186283275,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6251539301028084,
          "B_time": 22.177873930421732,
          "expected_cost": 0.0999999999999997,
          "final_B_probability": 0.3124950114859641,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 0,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.44192669861509914,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.5951351554687453,
          "B_time": 20.771020503156215,
          "expected_cost": 0.30000000000205906,
          "final_B_probability": 0.2944833984822441,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 0,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.7960474742204441,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.5466214850810984,
          "B_time": 18.616673353741895,
          "expected_cost": 0.9999999999998155,
          "final_B_probability": 0.26641030778558605,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 0,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.3597077041259236,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.49162887667130195,
          "B_time": 16.324969201589468,
          "expected_cost": 3.000000000000043,
          "final_B_probability": 0.23589706758902693,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 0,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.4540231564319225,
          "B_time": 14.83581715603145,
          "expected_cost": 6.606620883676392,
          "final_B_probability": 0.21571191458725283,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 0.25,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.29236681910959295,
          "delay": 0.25,
          "initial_state": "A",
          "first_B_probability": 0.6315737765240115,
          "B_time": 22.49347042982457,
          "expected_cost": 0.10000000000000035,
          "final_B_probability": 0.3164177793263769,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 0.25,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.502576419913529,
          "delay": 0.25,
          "initial_state": "A",
          "first_B_probability": 0.606699822507712,
          "B_time": 21.315740451722412,
          "expected_cost": 0.2999999999992391,
          "final_B_probability": 0.3013532858064232,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 0.25,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.9076063043826743,
          "delay": 0.25,
          "initial_state": "A",
          "first_B_probability": 0.5684462225714024,
          "B_time": 19.583689595478297,
          "expected_cost": 1.0000000000000167,
          "final_B_probability": 0.27887083703959487,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 0.25,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.5554481629884094,
          "delay": 0.25,
          "initial_state": "A",
          "first_B_probability": 0.5285485878702515,
          "B_time": 17.86631136064677,
          "expected_cost": 3.0000000000000133,
          "final_B_probability": 0.2561948806275964,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 0.25,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 0.25,
          "initial_state": "A",
          "first_B_probability": 0.5118712435085567,
          "B_time": 17.172379656512927,
          "expected_cost": 5.001893219495143,
          "final_B_probability": 0.24692419973721635,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 0.5,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.33128599142979726,
          "delay": 0.5,
          "initial_state": "A",
          "first_B_probability": 0.6370857384708568,
          "B_time": 22.765994518369013,
          "expected_cost": 0.09999999999994039,
          "final_B_probability": 0.31980837809368745,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 0.5,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.570176990936676,
          "delay": 0.5,
          "initial_state": "A",
          "first_B_probability": 0.616498270175608,
          "B_time": 21.782861535320617,
          "expected_cost": 0.2999999999999922,
          "final_B_probability": 0.30723916036577276,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 0.5,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.0318871798687275,
          "delay": 0.5,
          "initial_state": "A",
          "first_B_probability": 0.5864377501613421,
          "B_time": 20.399262188162833,
          "expected_cost": 1.0000000000011298,
          "final_B_probability": 0.2893342472410122,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 0.5,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.7731500690717834,
          "delay": 0.5,
          "initial_state": "A",
          "first_B_probability": 0.5577021888565581,
          "B_time": 19.128246765611703,
          "expected_cost": 3.0000000000000497,
          "final_B_probability": 0.2726643712200853,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 0.5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 0.5,
          "initial_state": "A",
          "first_B_probability": 0.5521857036060424,
          "B_time": 18.88954273312248,
          "expected_cost": 3.828008053728875,
          "final_B_probability": 0.2695100172470305,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 1,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.42271975062091205,
          "delay": 1,
          "initial_state": "A",
          "first_B_probability": 0.6459317369285825,
          "B_time": 23.206084852964402,
          "expected_cost": 0.10000000000046821,
          "final_B_probability": 0.3252939212364353,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 1,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.7289607990548377,
          "delay": 1,
          "initial_state": "A",
          "first_B_probability": 0.6319255661160775,
          "B_time": 22.528553991377393,
          "expected_cost": 0.29999999999944077,
          "final_B_probability": 0.316631422510994,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 1,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.3235430290765666,
          "delay": 1,
          "initial_state": "A",
          "first_B_probability": 0.6136556646572771,
          "B_time": 21.66613056872599,
          "expected_cost": 1.0000000000000138,
          "final_B_probability": 0.30551518678022455,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 1,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 1,
          "initial_state": "A",
          "first_B_probability": 0.6021793433705106,
          "B_time": 21.135986798993187,
          "expected_cost": 2.2988705565587773,
          "final_B_probability": 0.2986320067243724,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 1,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 1,
          "initial_state": "A",
          "first_B_probability": 0.6021793433705106,
          "B_time": 21.135986798993187,
          "expected_cost": 2.2988705565587773,
          "final_B_probability": 0.2986320067243724,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 2,
      "winner": 1,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 0.6728556092329888,
          "delay": 2,
          "initial_state": "A",
          "first_B_probability": 0.6575472685824733,
          "B_time": 23.78783269513876,
          "expected_cost": 0.09999999999999995,
          "final_B_probability": 0.33257975576046434,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 2,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.1631251507430198,
          "delay": 2,
          "initial_state": "A",
          "first_B_probability": 0.6513921039186027,
          "B_time": 23.485952854955553,
          "expected_cost": 0.2999999999999826,
          "final_B_probability": 0.3287087088419926,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 2,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 2,
          "initial_state": "A",
          "first_B_probability": 0.6457887544083064,
          "B_time": 23.213697695100272,
          "expected_cost": 0.8901568170562313,
          "final_B_probability": 0.32520647324190155,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 2,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 2,
          "initial_state": "A",
          "first_B_probability": 0.6457887544083064,
          "B_time": 23.213697695100272,
          "expected_cost": 0.8901568170562313,
          "final_B_probability": 0.32520647324190155,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 2,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 2,
          "initial_state": "A",
          "first_B_probability": 0.6457887544083064,
          "B_time": 23.213697695100272,
          "expected_cost": 0.8901568170562313,
          "final_B_probability": 0.32520647324190155,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 3,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.0427590523008,
          "delay": 3,
          "initial_state": "A",
          "first_B_probability": 0.6641838357456977,
          "B_time": 24.120857778008972,
          "expected_cost": 0.0999999999995716,
          "final_B_probability": 0.3367842629946062,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 3,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 1.8047118997345906,
          "delay": 3,
          "initial_state": "A",
          "first_B_probability": 0.6617771591067305,
          "B_time": 24.002367982240152,
          "expected_cost": 0.2999999999993739,
          "final_B_probability": 0.3352569298687346,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 3,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 3,
          "initial_state": "A",
          "first_B_probability": 0.6614020155370834,
          "B_time": 23.983939965234704,
          "expected_cost": 0.3685276850698048,
          "final_B_probability": 0.3350192067623082,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 3,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 3,
          "initial_state": "A",
          "first_B_probability": 0.6614020155370834,
          "B_time": 23.983939965234704,
          "expected_cost": 0.3685276850698048,
          "final_B_probability": 0.3350192067623082,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 3,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 3,
          "initial_state": "A",
          "first_B_probability": 0.6614020155370834,
          "B_time": 23.983939965234704,
          "expected_cost": 0.3685276850698048,
          "final_B_probability": 0.3350192067623082,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.1,
      "delay": 5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.0366330230251224,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6590623549795376,
          "B_time": 23.845719873895675,
          "expected_cost": 0.1000000000002591,
          "final_B_probability": 0.33352911190802065,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 5,
          "initial_state": "A",
          "first_B_probability": 0.6701703741146419,
          "B_time": 24.41953472557588,
          "expected_cost": 0.07302563835937853,
          "final_B_probability": 0.34060155664027086,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.06470198571745149,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.887174440173077,
          "expected_cost": 0.09999999999983533,
          "final_B_probability": 0.3298237523205622,
          "budget": 0.1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 0.3,
      "delay": 5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.06323081871211582,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6493849205407167,
          "B_time": 23.359542203125304,
          "expected_cost": 0.2999999999994145,
          "final_B_probability": 0.3274451388814204,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 5,
          "initial_state": "A",
          "first_B_probability": 0.6701703741146419,
          "B_time": 24.41953472557588,
          "expected_cost": 0.07302563835937853,
          "final_B_probability": 0.34060155664027086,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.11322639100722284,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 23.40054246200053,
          "expected_cost": 0.2999999999999927,
          "final_B_probability": 0.32068681325281606,
          "budget": 0.3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 1,
      "delay": 5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.11469533323039792,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6306245040485985,
          "B_time": 22.43854728614748,
          "expected_cost": 0.9999999999999805,
          "final_B_probability": 0.3158368169531022,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 5,
          "initial_state": "A",
          "first_B_probability": 0.6701703741146419,
          "B_time": 24.41953472557588,
          "expected_cost": 0.07302563835937853,
          "final_B_probability": 0.34060155664027086,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.21128257825898122,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 22.40126915696245,
          "expected_cost": 1.0000000000000098,
          "final_B_probability": 0.3022753356445355,
          "budget": 1,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 3,
      "delay": 5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.19671743249304982,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6007348114514977,
          "B_time": 21.02480784386656,
          "expected_cost": 3.0000000000481837,
          "final_B_probability": 0.29780643853319266,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 5,
          "initial_state": "A",
          "first_B_probability": 0.6701703741146419,
          "B_time": 24.41953472557588,
          "expected_cost": 0.07302563835937853,
          "final_B_probability": 0.34060155664027086,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.3813243057333883,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 20.631570025816277,
          "expected_cost": 3.000000000000078,
          "final_B_probability": 0.27083533749065825,
          "budget": 3,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    },
    {
      "budget": 10,
      "delay": 5,
      "winner": 0,
      "results": [
        {
          "policy": "减少起步",
          "u": 0.3531077761175254,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.54438199163591,
          "B_time": 18.514055729586737,
          "expected_cost": 10.00000000013951,
          "final_B_probability": 0.26515522185045143,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "改变通道",
          "u": 2.0,
          "delay": 5,
          "initial_state": "A",
          "first_B_probability": 0.6701703741146419,
          "B_time": 24.41953472557588,
          "expected_cost": 0.07302563835937853,
          "final_B_probability": 0.34060155664027086,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        },
        {
          "policy": "促进返回",
          "u": 0.7793368837988413,
          "delay": 0,
          "initial_state": "A",
          "first_B_probability": 0.6723538280186192,
          "B_time": 16.464537617882396,
          "expected_cost": 10.000000000000261,
          "final_B_probability": 0.20271367943388346,
          "budget": 10,
          "stages": 16,
          "accuracy": 1,
          "false_A": 0
        }
      ]
    }
  ],
  "routing_sensor_sensitivity": [
    {
      "policy": "改变通道",
      "u": 0.7960474742204441,
      "delay": 0,
      "initial_state": "A",
      "first_B_probability": 0.5466214850810984,
      "B_time": 18.616673353741895,
      "expected_cost": 0.9999999999998155,
      "final_B_probability": 0.26641030778558605,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 1.3235430290765666,
      "delay": 1,
      "initial_state": "A",
      "first_B_probability": 0.6136556646572771,
      "B_time": 21.66613056872599,
      "expected_cost": 1.0000000000000138,
      "final_B_probability": 0.30551518678022455,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 2.0,
      "delay": 3,
      "initial_state": "A",
      "first_B_probability": 0.6614020155370834,
      "B_time": 23.983939965234704,
      "expected_cost": 0.3685276850698048,
      "final_B_probability": 0.3350192067623082,
      "budget": 1,
      "stages": 16,
      "accuracy": 1,
      "false_A": 0
    },
    {
      "policy": "改变通道",
      "u": 0.715025223837696,
      "delay": 1,
      "initial_state": "A",
      "first_B_probability": 0.6408131133459397,
      "B_time": 22.95676140462919,
      "expected_cost": 0.9999999999967135,
      "final_B_probability": 0.3221134467972953,
      "budget": 1,
      "stages": 16,
      "accuracy": 0.8,
      "false_A": 0.02
    },
    {
      "policy": "改变通道",
      "u": 0.8019985930145935,
      "delay": 3,
      "initial_state": "A",
      "first_B_probability": 0.6668066264944807,
      "B_time": 24.25051976084518,
      "expected_cost": 0.9999999999999764,
      "final_B_probability": 0.338453233079547,
      "budget": 1,
      "stages": 16,
      "accuracy": 0.8,
      "false_A": 0.02
    }
  ],
  "delay_stage_convergence": [
    {
      "stages": 4,
      "survive_preparation": 0.10662224073302788,
      "result": {
        "policy": "改变通道",
        "u": 2.0,
        "delay": 3,
        "initial_state": "A",
        "first_B_probability": 0.6539134271970153,
        "B_time": 23.608444571673576,
        "expected_cost": 0.6219709676241738,
        "final_B_probability": 0.33029085127240443,
        "budget": 1,
        "stages": 4,
        "accuracy": 1,
        "false_A": 0
      }
    },
    {
      "stages": 8,
      "survive_preparation": 0.07826695083372824,
      "result": {
        "policy": "改变通道",
        "u": 2.0,
        "delay": 3,
        "initial_state": "A",
        "first_B_probability": 0.6589072850016245,
        "B_time": 23.858571717850346,
        "expected_cost": 0.4531591710390285,
        "final_B_probability": 0.33343966013391874,
        "budget": 1,
        "stages": 8,
        "accuracy": 1,
        "false_A": 0
      }
    },
    {
      "stages": 16,
      "survive_preparation": 0.06395317456515288,
      "result": {
        "policy": "改变通道",
        "u": 2.0,
        "delay": 3,
        "initial_state": "A",
        "first_B_probability": 0.6614020155370834,
        "B_time": 23.983939965234704,
        "expected_cost": 0.3685276850698048,
        "final_B_probability": 0.3350192067623082,
        "budget": 1,
        "stages": 16,
        "accuracy": 1,
        "false_A": 0
      }
    },
    {
      "stages": 32,
      "survive_preparation": 0.056835788180370536,
      "result": {
        "policy": "改变通道",
        "u": 2.0,
        "delay": 3,
        "initial_state": "A",
        "first_B_probability": 0.6626360911543063,
        "B_time": 24.046021300398706,
        "expected_cost": 0.32660701433176853,
        "final_B_probability": 0.3358021578620651,
        "budget": 1,
        "stages": 32,
        "accuracy": 1,
        "false_A": 0
      }
    },
    {
      "stages": 64,
      "survive_preparation": 0.053300460610330426,
      "result": {
        "policy": "改变通道",
        "u": 2.0,
        "delay": 3,
        "initial_state": "A",
        "first_B_probability": 0.6632475173169333,
        "B_time": 24.076788330951757,
        "expected_cost": 0.3058272042877015,
        "final_B_probability": 0.3361904591716636,
        "budget": 1,
        "stages": 64,
        "accuracy": 1,
        "false_A": 0
      }
    }
  ],
  "hard_budget_trials": [
    {
      "case": 0,
      "policy": "减少起步",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.1,
      "training_scores": [
        0.6555,
        0.6415,
        0.6225,
        0.6585,
        0.6695,
        0.669,
        0.6705,
        0.668,
        0.6695,
        0.67,
        0.671,
        0.671
      ],
      "test_first_B": 0.628,
      "first_B_SE": 0.003946572989640131,
      "test_B_time": 22.339999595366237,
      "B_time_SE": 0.21407954323319692,
      "spent_mean": 0.7611707763902182,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.16913333333333333,
      "paired_change_first": -0.03746666666666667,
      "paired_change_first_SE": 0.001550598162220374,
      "paired_change_B_time": -1.838549267254109,
      "paired_change_B_time_SE": 0.023218839503986927,
      "baseline_first_B": 0.6654666666666667,
      "baseline_B_time": 24.178548862620286,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 0,
      "policy": "改变通道",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.65,
      "training_scores": [
        0.6555,
        0.649,
        0.641,
        0.6265,
        0.6095,
        0.592,
        0.5785,
        0.589,
        0.608,
        0.6175,
        0.624,
        0.628
      ],
      "test_first_B": 0.586,
      "first_B_SE": 0.004021775517342103,
      "test_B_time": 20.25809758442866,
      "B_time_SE": 0.20837339025518203,
      "spent_mean": 0.4975489495907048,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.24106666666666668,
      "paired_change_first": -0.07946666666666667,
      "paired_change_first_SE": 0.0029070049447647277,
      "paired_change_B_time": -3.92045127819158,
      "paired_change_B_time_SE": 0.15907682084214866,
      "baseline_first_B": 0.6654666666666667,
      "baseline_B_time": 24.178548862620286,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 0,
      "policy": "促进返回",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.0,
      "training_scores": [
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555,
        0.6555
      ],
      "test_first_B": 0.6654666666666667,
      "first_B_SE": 0.003852580300674321,
      "test_B_time": 24.178548862620275,
      "B_time_SE": 0.21778433509157635,
      "spent_mean": 0.0,
      "spent_max": 0.0,
      "fraction_budget_exhausted": 0.0,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": 0.0,
      "paired_change_B_time_SE": 0.0,
      "baseline_first_B": 0.6654666666666667,
      "baseline_B_time": 24.178548862620286,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 1,
      "policy": "减少起步",
      "budget": 1,
      "delay": 3,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.1,
      "training_scores": [
        0.681,
        0.659,
        0.638,
        0.655,
        0.6655,
        0.6675,
        0.665,
        0.669,
        0.67,
        0.671,
        0.6695,
        0.67
      ],
      "test_first_B": 0.6412,
      "first_B_SE": 0.003916443101646419,
      "test_B_time": 22.61890740001607,
      "B_time_SE": 0.21284681865542177,
      "spent_mean": 0.7582066803401523,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.16313333333333332,
      "paired_change_first": -0.0356,
      "paired_change_first_SE": 0.0015129425852790947,
      "paired_change_B_time": -1.8502957107508862,
      "paired_change_B_time_SE": 0.023216373965382738,
      "baseline_first_B": 0.6768,
      "baseline_B_time": 24.469203110767005,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 1,
      "policy": "改变通道",
      "budget": 1,
      "delay": 3,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.65,
      "training_scores": [
        0.681,
        0.6845,
        0.6845,
        0.684,
        0.6825,
        0.68,
        0.6795,
        0.68,
        0.681,
        0.6815,
        0.6835,
        0.6825
      ],
      "test_first_B": 0.6716,
      "first_B_SE": 0.00383465030602929,
      "test_B_time": 24.201026479632887,
      "B_time_SE": 0.21638064327740256,
      "spent_mean": 0.0345323267053446,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.0084,
      "paired_change_first": -0.0052,
      "paired_change_first_SE": 0.001365645756613297,
      "paired_change_B_time": -0.26817663113403967,
      "paired_change_B_time_SE": 0.07976360834636335,
      "baseline_first_B": 0.6768,
      "baseline_B_time": 24.469203110767005,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 1,
      "policy": "促进返回",
      "budget": 1,
      "delay": 3,
      "initial_state": "A",
      "objective": "first",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.0,
      "training_scores": [
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681,
        0.681
      ],
      "test_first_B": 0.6768,
      "first_B_SE": 0.0038188684515723646,
      "test_B_time": 24.46920311076696,
      "B_time_SE": 0.21668776965025277,
      "spent_mean": 0.0,
      "spent_max": 0.0,
      "fraction_budget_exhausted": 0.0,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": 0.0,
      "paired_change_B_time_SE": 0.0,
      "baseline_first_B": 0.6768,
      "baseline_B_time": 24.469203110767005,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 2,
      "policy": "减少起步",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.1,
      "training_scores": [
        24.796722101264045,
        23.921046333043225,
        23.06394662108223,
        23.377814666860495,
        23.78064688933072,
        24.262457712690892,
        24.34263431445855,
        24.39204973551389,
        24.547120624222156,
        24.6425463715639,
        24.683554904961493,
        24.700740964410556
      ],
      "test_first_B": 0.6319333333333333,
      "first_B_SE": 0.003937927566872575,
      "test_B_time": 22.852743701034402,
      "B_time_SE": 0.21579309281586015,
      "spent_mean": 0.7559404734621709,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.1656,
      "paired_change_first": -0.03726666666666666,
      "paired_change_first_SE": 0.0015466146738209268,
      "paired_change_B_time": -1.7608059866541592,
      "paired_change_B_time_SE": 0.022634552474844905,
      "baseline_first_B": 0.6692,
      "baseline_B_time": 24.613549687688526,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 2,
      "policy": "改变通道",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.65,
      "training_scores": [
        24.796722101264045,
        24.410053583337618,
        23.815362933986023,
        23.073386749456485,
        22.098926540003138,
        21.121804331605656,
        20.42896757832598,
        20.541180011713543,
        21.900856098152705,
        22.57313488174357,
        22.79790485131666,
        23.02779589179373
      ],
      "test_first_B": 0.5881333333333333,
      "first_B_SE": 0.004018695190121515,
      "test_B_time": 20.60498524985604,
      "B_time_SE": 0.20893117671045844,
      "spent_mean": 0.49910112350199076,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.23913333333333334,
      "paired_change_first": -0.08106666666666666,
      "paired_change_first_SE": 0.0028732844309721953,
      "paired_change_B_time": -4.00856443783255,
      "paired_change_B_time_SE": 0.1604850045646055,
      "baseline_first_B": 0.6692,
      "baseline_B_time": 24.613549687688526,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 2,
      "policy": "促进返回",
      "budget": 1,
      "delay": 0,
      "initial_state": "A",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.2,
      "training_scores": [
        24.796722101264045,
        24.29820732922052,
        23.80133439034554,
        23.576522717117626,
        23.729393480377425,
        24.05253788036014,
        24.274342920505486,
        24.2961127975468,
        24.340184112769705,
        24.360918444569535,
        24.334320829629174,
        24.337219989967014
      ],
      "test_first_B": 0.6692,
      "first_B_SE": 0.003841754105219308,
      "test_B_time": 23.227099923601354,
      "B_time_SE": 0.21311739275290342,
      "spent_mean": 0.5186195089804073,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.378,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": -1.3864497640870987,
      "paired_change_B_time_SE": 0.0968384367261527,
      "baseline_first_B": 0.6692,
      "baseline_B_time": 24.613549687688526,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 3,
      "policy": "减少起步",
      "budget": 1,
      "delay": 0,
      "initial_state": "B",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.3,
      "training_scores": [
        56.53780434316862,
        56.06490143475017,
        55.59186821193394,
        54.93968552536585,
        54.77275028274114,
        55.05367350999264,
        55.294334116202066,
        55.429555364036034,
        55.50011720450339,
        55.5115011239879,
        55.51247165761454,
        55.53060723638653
      ],
      "test_first_B": 1.0,
      "first_B_SE": 0.0,
      "test_B_time": 54.83573086678748,
      "B_time_SE": 0.2478783019575802,
      "spent_mean": 0.8347783931615933,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.8019333333333334,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": -0.5019845492618472,
      "paired_change_B_time_SE": 0.180352245058754,
      "baseline_first_B": 1.0,
      "baseline_B_time": 55.337715416049214,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 3,
      "policy": "改变通道",
      "budget": 1,
      "delay": 0,
      "initial_state": "B",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.9,
      "training_scores": [
        56.53780434316862,
        56.18840152934434,
        55.986119127644024,
        55.565025847670164,
        55.03948405215869,
        54.66049541664081,
        54.51092423254066,
        54.388389427185174,
        54.71547275780303,
        55.03743777821469,
        55.29659836625819,
        55.420484758329096
      ],
      "test_first_B": 1.0,
      "first_B_SE": 0.0,
      "test_B_time": 53.615329379534614,
      "B_time_SE": 0.25384896991395645,
      "spent_mean": 0.421748677443673,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.2816666666666667,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": -1.7223860365146675,
      "paired_change_B_time_SE": 0.13300978602999697,
      "baseline_first_B": 1.0,
      "baseline_B_time": 55.337715416049214,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    },
    {
      "case": 3,
      "policy": "促进返回",
      "budget": 1,
      "delay": 0,
      "initial_state": "B",
      "objective": "occupation",
      "candidate_amplitudes": [
        0.0,
        0.05,
        0.1,
        0.2,
        0.3,
        0.45,
        0.65,
        0.9,
        1.2,
        1.5,
        1.75,
        2.0
      ],
      "train_n": 2000,
      "test_n": 15000,
      "chosen_u": 0.3,
      "training_scores": [
        56.53780434316862,
        55.17908517757022,
        53.80259565671551,
        53.0012907297484,
        52.66855749494879,
        53.11274820971582,
        53.490671585472214,
        53.88038545194754,
        53.97127324931975,
        53.874704203371174,
        53.91235677859765,
        53.89059026137488
      ],
      "test_first_B": 1.0,
      "first_B_SE": 0.0,
      "test_B_time": 53.06920063529682,
      "B_time_SE": 0.2553486692548109,
      "spent_mean": 0.9450694535107326,
      "spent_max": 1.0,
      "fraction_budget_exhausted": 0.8893333333333333,
      "paired_change_first": 0.0,
      "paired_change_first_SE": 0.0,
      "paired_change_B_time": -2.2685147807526684,
      "paired_change_B_time_SE": 0.2853209892389546,
      "baseline_first_B": 1.0,
      "baseline_B_time": 55.337715416049214,
      "scope": "best of finite constant-amplitude/state-gated candidates, NOT globally optimal feedback"
    }
  ],
  "limits": "expected resource atlas distinct from hard per-path cap; finite candidate family, restricted rate actuators; no original molecular/force optimum"
}