【 李宏毅深度学习合辑 】Advanced Topics in Deep Learning - Imitation Learning

 

 

 

 

 

 

 

 

You have to force experts to treat some uncommon and extreme situations.

 

 

 

 

 

 

 

 

a mechanical way to learn

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

However, we don't know r_t

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 if you use sequence GAN, you could get different D(tao_i) in different steps. However here, it remains the same.