Consistency-Guided Temperature Scaling Using Style and Content Information for Out-of-Domain Calibration

arXiv:2402.15019v1 Announce Type: new
Abstract: Research interests in the robustness of deep neural networks against domain shifts have been rapidly increasing in recent years. Most existing works, however, focus on improving the accuracy of the model, not the calibration performance which is another important requirement for trustworthy AI systems. Temperature scaling (TS), an accuracy-preserving post-hoc calibration method, has been proven to be effective in in-domain settings, but not in out-of-domain (OOD) due to the difficulty in obtaining a validation set …

abstract accuracy ai systems arxiv cs.ai cs.lg domain focus information networks neural networks performance research robustness scaling stat.ml style systems trustworthy trustworthy ai type