learning_rate: 0.0003
learning_rate_schedule: constant
batch_size: 64
buffer_size: 12000
buffer_size: 200000
buffer_init_steps: 0
tau: 0.005
steps_per_update: 10.0
gamma: 0.99
strength: 1.0
keep_checkpoints: 5
max_steps: 500000
max_steps: 200000
time_horizon: 1000
summary_freq: 12000
threaded: true
batch_size: 256
buffer_size: 50000
buffer_size: 500000