Does this really work???

[garymcintire]

Looking at the code here, I do not see the synthetic gradients module.
dni_linear seems to just be a regular MLP with batch normalization
Am I missing something???

[andrewliao11]

Hi @garymcintire ,
In fact, my dni_linear is the synthetic gradient module. The main network will be updated based on the output of the dni_linear, instead of back-propagation signal from loss function.
please see L37 in train.py.

[andrewliao11]

I'll first close this issue.
feel free to reopen it if there are other questions

[garymcintire]

Thanks for clearing this up, Andrew!

…

-----Original Message----- From: Andrew [mailto:notifications@github.com] Sent: Thursday, July 20, 2017 3:37 AM To: andrewliao11/dni.pytorch Cc: garymcintire; Mention Subject: Re: [andrewliao11/dni.pytorch] Does this really work??? (#4) Closed #4 <#4> . — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#4 (comment)> , or mute the thread <https://github.com/notifications/unsubscribe-auth/AE_8eHCBxmNVTPdI3MsTZVBtqrIORtawks5sPxGZgaJpZM4OdW30> . <https://github.com/notifications/beacon/AE_8eDKP064nb8gGpXvBerA1dXzVl5Cmks5sPxGZgaJpZM4OdW30.gif>