• Home
  • Search
  • Generality-Training of a Classifier for Improved Calibration in Unseen Contexts
  • https://doi.org/10.1007/978-3-031-43424-2_23Copy DOI Icon

Generality-Training of a Classifier for Improved Calibration in Unseen Contexts

  • Jan 1, 2023
  • Bhawani Shankar Leelar +1 more
Show More
  • Abstract
  • PDF
  • Literature Map
  • References
  • Similar Papers
Abstract

Abstract Artificial neural networks tend to output class probabilities that are miscalibrated, i.e., their reported uncertainty is not a very good indicator of how much we should trust the model. Consequently, methods have been developed to improve the model’s predictive uncertainty, both during training and post-hoc. Even if the model is calibrated on the domain used in training, it typically becomes over-confident when applied on slightly different target domains, e.g. due to perturbations or shifts in the data. The model can be recalibrated for a fixed list of target domains, but its performance can still be poor on unseen target domains. To address this issue, we propose a generality-training procedure that learns a modified head for the neural network to achieve better calibration generalization to new domains while retaining calibration performance on the given domains. This generality-head is trained on multiple domains using a new objective function with increased emphasis on the calibration loss compared to cross-entropy. Such training results in a more general model in the sense of not only better calibration but also better accuracy on unseen domains, as we demonstrate experimentally on multiple datasets. The code and supplementary for the paper is available (https://github.com/bsl-traveller/CaliGen.git).

Loading PDF

Similar Papers
  • Research Article

Frequency-aware domain randomization for single-source domain generalization in medical image segmentation.

  • Nov 29, 2025
  • Medical physics
  • Jiayi Wu +6
  • Conference Article
  • Citations61

Adaptive Mixture of Experts Learning for Generalizable Face Anti-Spoofing

  • Oct 10, 2022
  • Qianyu Zhou +5
  • Book Chapter
  • Citations51

Mimic Embedding via Adaptive Aggregation: Learning Generalizable Person Re-identification

  • Jan 01, 2022
  • Boqiang Xu +3
  • Research Article
  • Citations22

Dynamic domain generalization for medical image segmentation.

  • Apr 01, 2025
  • Neural networks : the official journal of the International Neural Network Society
  • Zhiming Cheng +3
  • Conference Article
  • Citations3

Domain Camera Adaptation and Collaborative Multiple Feature Clustering for Unsupervised Person Re-ID

  • Oct 10, 2022
  • Yuanpeng Tu
  • Research Article
  • Citations35

Generalizable model-agnostic semantic segmentation via target-specific normalization

  • Sep 08, 2021
  • Pattern Recognition
  • Jian Zhang +3
  • Research Article
  • Citations13

Domain-Aware Dual Attention for Generalized Medical Image Segmentation on Unseen Domains.

  • May 01, 2023
  • IEEE Journal of Biomedical and Health Informatics
  • Huilin Lai +4
  • Research Article
  • Citations4

Phrase Grounding-Based Style Transfer for Single-Domain Generalized Object Detection

  • Jan 01, 2026
  • IEEE Transactions on Circuits and Systems for Video Technology
  • Hao Li +8
  • Research Article
  • Citations2

Adaptive wavelet‐VNet for single‐sample test time adaptation in medical image segmentation

  • Oct 01, 2024
  • Medical Physics
  • Xiaoxue Qian +2
  • Research Article
  • Citations7

Adaptive Domain Generalization Via Online Disagreement Minimization.

  • Jan 01, 2023
  • IEEE Transactions on Image Processing
  • Xin Zhang +1
  • Research Article
  • Citations12

Multi-Source Collaborative Gradient Discrepancy Minimization for Federated Domain Generalization

  • Mar 24, 2024
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Yikang Wei +1
  • Conference Article
  • Citations43

Generalization on Unseen Domains via Inference-time Label-Preserving Target Projections

  • Jun 01, 2021
  • Prashant Pandey +3
  • Conference Article
  • Citations1

Discovery of New Multi-Level Features for Domain Generalization via Knowledge Corruption

  • Aug 21, 2022
  • Ahmed Frikha +2
  • Research Article

Federated Multi-source Domain Adaptation via Contrastive Cross-domain Semantic Alignment with Adversarial Feature Augmentation

  • May 16, 2026
  • ACM Transactions on Knowledge Discovery from Data
  • Wenjie Mao +3
  • Research Article
  • Citations7

UniAda: Domain Unifying and Adapting Network for Generalizable Medical Image Segmentation.

  • May 01, 2025
  • IEEE transactions on medical imaging
  • Zhongzhou Zhang +7
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.