Why is the learning rate for the bias usually twice as large as the the LR for the weights?

Why is the learning rate for the bias usually twice as large as the the LR for the weights?

[Greg Heath]

Please post the proof of this assertion.

Greg

[Salma Hassan]

https://datascience.stackexchange.com/questions/23549/why-is-the-learning-rate-for-the-bias-usually-twice-as-large-as-the-the-lr-for-t

on the other hand if you check, for example alexnet layers you will find that net.Layers(N, 1).BiasLearnRateFactor=2 and net.Layers(N, 1).WeightLearnRateFactor = 1