curiosity_strength: 0.01
curiosity_enc_size: 128
BananaBrain:
BananaLearning:
normalize: false
batch_size: 1024
beta: 5.0e-3
BouncerBrain:
BouncerLearning:
PushBlockBrain:
PushBlockLearning:
max_steps: 5.0e4
batch_size: 128
buffer_size: 2048
time_horizon: 64
num_layers: 2
SmallWallBrain:
SmallWallLearning:
max_steps: 1.0e6
BigWallBrain:
BigWallLearning:
StrikerBrain:
StrikerLearning:
max_steps: 5.0e5
learning_rate: 1e-3
GoalieBrain:
GoalieLearning:
batch_size: 320
PyramidBrain:
PyramidLearning:
use_curiosity: true
summary_freq: 2000
num_epoch: 3
VisualPyramidBrain:
VisualPyramidLearning:
curiosity_enc_size: 256
Ball3DBrain:
Ball3DLearning:
normalize: true
batch_size: 64
buffer_size: 12000
gamma: 0.995
beta: 0.001
Ball3DHardBrain:
Ball3DHardLearning:
batch_size: 1200
TennisBrain:
TennisLearning:
CrawlerBrain:
CrawlerLearning:
time_horizon: 1000
num_layers: 3
hidden_units: 512
WalkerBrain:
WalkerLearning:
ReacherBrain:
ReacherLearning:
max_steps: 1e6
summary_freq: 3000
HallwayBrain:
HallwayLearning:
use_recurrent: true
sequence_length: 64
summary_freq: 1000
VisualHallwayBrain:
VisualHallwayLearning:
num_layers: 1
VisualPushBlockBrain:
VisualPushBlockLearning:
sequence_length: 32
GridWorldBrain:
GridWorldLearning:
batch_size: 32
time_horizon: 5
BasicBrain:
BasicLearning: