Calibration of Ordinal Regression Networks

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kim, Daehwan, Chung, Haejun, Jang, Ikbeom
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910811618803712
author Kim, Daehwan
Chung, Haejun
Jang, Ikbeom
author_facet Kim, Daehwan
Chung, Haejun
Jang, Ikbeom
contents Recent studies have shown that deep neural networks are not well-calibrated and often produce over-confident predictions. The miscalibration issue primarily stems from using cross-entropy in classifications, which aims to align predicted softmax probabilities with one-hot labels. In ordinal regression tasks, this problem is compounded by an additional challenge: the expectation that softmax probabilities should exhibit unimodal distribution is not met with cross-entropy. The ordinal regression literature has focused on learning orders and overlooked calibration. To address both issues, we propose a novel loss function that introduces ordinal-aware calibration, ensuring that prediction confidence adheres to ordinal relationships between classes. It incorporates soft ordinal encoding and ordinal-aware regularization to enforce both calibration and unimodality. Extensive experiments across four popular ordinal regression benchmarks demonstrate that our approach achieves state-of-the-art calibration without compromising classification accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15658
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Calibration of Ordinal Regression Networks
Kim, Daehwan
Chung, Haejun
Jang, Ikbeom
Machine Learning
Computer Vision and Pattern Recognition
Recent studies have shown that deep neural networks are not well-calibrated and often produce over-confident predictions. The miscalibration issue primarily stems from using cross-entropy in classifications, which aims to align predicted softmax probabilities with one-hot labels. In ordinal regression tasks, this problem is compounded by an additional challenge: the expectation that softmax probabilities should exhibit unimodal distribution is not met with cross-entropy. The ordinal regression literature has focused on learning orders and overlooked calibration. To address both issues, we propose a novel loss function that introduces ordinal-aware calibration, ensuring that prediction confidence adheres to ordinal relationships between classes. It incorporates soft ordinal encoding and ordinal-aware regularization to enforce both calibration and unimodality. Extensive experiments across four popular ordinal regression benchmarks demonstrate that our approach achieves state-of-the-art calibration without compromising classification accuracy.
title Calibration of Ordinal Regression Networks
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.15658