TensorFlow_RLRE icon indicating copy to clipboard operation
TensorFlow_RLRE copied to clipboard

why do you update parameters manually?

Open FrankWork opened this issue 6 years ago • 1 comments

FrankWork avatar Jun 05 '18 13:06 FrankWork

To fit the parameters' updating rule in the algorithm 2 of the paper: para_old = update_rate* para_new + (1-update_rate)* para_old

xuyanfu avatar Jun 05 '18 14:06 xuyanfu