Different definitions of "Loss of Supervised Learning" between the paper and the code

[wuweijia1994]

In the paper, we could see the loss during the supervising learning should be the KL divergence of between the trajectories and the neural network output:

However, in the code, why it becomes the || label - output ||2?

Is it because of the convenience of implementation?

[dujinyu]

I have many questions to consult you, can you add QQ 1242642280? @wuweijia1994