You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently MapObjective is a nice idea, but has some flaws
KLObjective is not really measuring the KL, it's off by a constant. Maybe we just disregard this and call it a quirk.
In practice, we need regularization. It would be really cool to have another abstract class like RegularizedObjective::MapObjective, where we create an abstract class RegularizerBase with e.g. child class L2Regularizer(lambda). Then, we could do something like RegularizedObjective obj = KLObjective(train, test, dim) + L2Regularizer(0.05).
The text was updated successfully, but these errors were encountered:
Currently
MapObjective
is a nice idea, but has some flawsKLObjective
is not really measuring the KL, it's off by a constant. Maybe we just disregard this and call it a quirk.RegularizedObjective::MapObjective
, where we create an abstract classRegularizerBase
with e.g. child classL2Regularizer(lambda)
. Then, we could do something likeRegularizedObjective obj = KLObjective(train, test, dim) + L2Regularizer(0.05)
.The text was updated successfully, but these errors were encountered: