MAGDiff: Covariate Data Set Shift Detection via Activation Graphs of Deep Neural Networks

arXiv:2305.13271v2 Announce Type: replace-cross
Abstract: Despite their successful application to a variety of tasks, neural networks remain limited, like other machine learning methods, by their sensitivity to shifts in the data: their performance can be severely impacted by differences in distribution between the data on which they were trained and that on which they are deployed. In this article, we propose a new family of representations, called MAGDiff, that we extract from any given neural network classifier and that allows …

abstract application arxiv cs.lg data data set detection differences distribution graphs machine machine learning networks neural networks performance replace sensitivity set shift stat.ml tasks type via