Error-Driven Uncertainty Aware Training

arXiv:2405.01205v1 Announce Type: new
Abstract: Neural networks are often overconfident about their predictions, which undermines their reliability and trustworthiness. In this work, we present a novel technique, named Error-Driven Uncertainty Aware Training (EUAT), which aims to enhance the ability of neural models to estimate their uncertainty correctly, namely to be highly uncertain when they output inaccurate predictions and low uncertain when their output is accurate. The EUAT approach operates during the model's training phase by selectively employing two loss functions …

abstract arxiv cs.cv cs.lg error networks neural networks novel predictions reliability training type uncertain uncertainty work