markovian_ppo/config.json at dev · ucsb/markovian_ppo · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
{
  "environments": {
    "CartPole-v1": {
      "time_steps": 40,
      "num_trials": 2,
      "n_envs": 1
    },
    "Pendulum-v1": {
      "time_steps": 100,
      "num_trials": 2,
      "n_envs": 1
    },
    "MountainCarContinuous-v0": {
      "time_steps": 100,
      "num_trials": 2,
      "n_envs": 1
    }
  },
  "noise_strategies": {
    "gaussian": [
      {
        "mean": 0,
        "std": 0.002
      },
      {
        "mean": 0,
        "std": 0.004
      },
      {
        "mean": 0,
        "std": 0.006
      },
      {
        "mean": 0,
        "std": 0.008
      },
      {
        "mean": 0,
        "std": 0.01
      }
    ],
    "uniform": [
      {
        "low": -0.05,
        "high": 0.05
      },
      {
        "low": -0.1,
        "high": 0.1
      },
      {
        "low": -0.2,
        "high": 0.2
      }
    ],
    "salt_and_pepper": [
      {
        "prob": 0.05
      },
      {
        "prob": 0.1
      },
      {
        "prob": 0.2
      }
    ]
  }
}