search

yaringal / ConcreteDropout

Code for Concrete Dropout as presented in https://arxiv.org/abs/1705.07832
MIT License
245 stars 68 forks source link

value to set for weight_regularizer & dropout_regularizer when dataset size is unknown #12

Open nikhil-salodkar opened 4 years ago

nikhil-salodkar commented 4 years ago

Hi,

To calculate the value of weight_regularizer & dropout_regularizer the dataset_size N is used. What if we don't know the dataset size in advance like in case of Lifelong Reinforcement learning in which case re-training happens on a periodic basis on new data as it comes.

I am not able to figure out in such cases how we will be able to use this concrete dropout implementation.

Can you please suggest on how to handle such situations using Concrete dropout. Will really appreciate it.

Thanks in Advance Nikhil

JanRuettinger commented 4 years ago

I am facing the same issue. Did you find an answer?

yaringal commented 4 years ago

You can update N as the amount of data increases (have a look at this). In effect this will push the dropout p towards 0 in the limit of data