{
    "experiment": "A",
    "trainenv": {
        "stop_on_collision": false
    },
    "policy": {
        "learning_rate": 0.0003,
        "learning_rate_decay": 1.0,
        "delta": 0.01,
        "iterations_per_epoch": 100,
        "hidden_layer_size": 40,
        "n_hidden_layers": 2,
        "activation": 0
    },
    "value": {
        "learning_rate": 0.0001,
        "weight_decay": 0.001,
        "iterations_per_epoch": 1000
    },
    "discriminator": {
        "learning_rate": 0.001,
        "weight_decay": 0.0001,
        "iterations_per_epoch": 100,
        "n_hidden_layers_element": 4,
        "n_hidden_layers_global": 1,
        "hidden_layer_size": 10,
        "activation": 0
    },
    "train_epochs": 100,
    "seed": 0
}