bindsnet
bindsnet copied to clipboard
Learning rate decay
We've no in-library implementation of learning rate decay. It could be implemented similarly to weight decay in Connection
objects, but in LearningRule
objects, operating on nu
.
Hello, sorry if this is an erroneous question, I am just starting to learn about this library.
Would it make sense for this feature to implement the option of defining individual learning rate decays to operate on pre and post synaptic events in addition to operating on both of them with the same value?
Yes, it may help, but for that we need to find the relevant biological evidence for this separate adjustment and also application that can utilize it.