yangmuzhi / airl

learning robust rewards with adversarial inverse reinforcement learning

Airl algo

• trajectory/ storage trajectories of expert like (s,a)

Useage

python run.py

Result

rewards:

dscriminator loss:

Algo

Reference paper

learning robust rewards with adversarial inverse reinforcement learning