{
  "true_transition": [
    [
      1,
      0
    ],
    [
      0,
      1
    ]
  ],
  "model_transition": [
    [
      1,
      0
    ],
    [
      0,
      1
    ]
  ],
  "rewards": [
    1,
    2
  ],
  "discount": 0.5,
  "horizon": 4
}