Approximate Bayesian Class-Conditional Models under Continuous Representation Shift

arXiv:2305.19076v2 Announce Type: replace
Abstract: For models consisting of a classifier in some representation space, learning online from a non-stationary data stream often necessitates changes in the representation. So, the question arises of what is the best way to adapt the classifier to shifts in representation. Current methods only slowly change the classifier to representation shift, introducing noise into learning as the classifier is misaligned to the representation. We propose DeepCCG, an empirical Bayesian approach to solve this problem. DeepCCG …

abstract adapt arxiv bayesian change class classifier continuous cs.lg current data data stream question representation shift space stat.ml type