Decoupling Feature Extraction and Classification Layers for Calibrated Neural Networks

arXiv:2405.01196v1 Announce Type: new
Abstract: Deep Neural Networks (DNN) have shown great promise in many classification applications, yet are widely known to have poorly calibrated predictions when they are over-parametrized. Improving DNN calibration without comprising on model accuracy is of extreme importance and interest in safety critical applications such as in the health-care sector. In this work, we show that decoupling the training of feature extraction layers and classification layers in over-parametrized DNN architectures such as Wide Residual Networks (WRN) …

abstract accuracy applications arxiv calibration classification cs.lg dnn extraction feature feature extraction importance improving model accuracy networks neural networks predictions safety stat.ml type