# Calibrating Where It Matters: Constrained Temperature Scaling

arXiv: https://arxiv.org/abs/2406.11456 · HTML: https://mlsanity.com/paper/2406.11456/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 17 Jun 2024.

## Builds on (most cited first, 9 of 9)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1710.05006/) (2017)
- [Revisiting the Calibration of Modern Neural Networks](https://mlsanity.com/paper/2106.07998/) (2021)
- [BCN20000: Dermoscopic Lesions in the Wild](https://mlsanity.com/paper/1908.02288/) (2019)
- [Calibrating Predictions to Decisions: A Novel Approach to Multi-Class Calibration](https://mlsanity.com/paper/2107.05719/) (2021)
- [Classifier Calibration: A survey on how to assess and improve predicted class probabilities](https://mlsanity.com/paper/2112.10327/) (2021)

## Built on by (most cited first, 2 of 2)

- [CoV: Chain-of-View Prompting for Spatial Reasoning](https://mlsanity.com/paper/2601.05172/) (2026)
- [Optimizing for the decision not the prediction: an exploration of Smooth Net Benefit as a training objective](https://mlsanity.com/paper/2609.12752/) (2026)

## Abstract (arXiv)

We consider calibration of convolutional classifiers for diagnostic decision making. Clinical decision makers can use calibrated classifiers to minimise expected costs given their own cost function. Such functions are usually unknown at training time. If minimising expected costs is the primary aim, algorithms should focus on tuning calibration in regions of probability simplex likely to effect decisions. We give an example, modifying temperature scaling calibration, and demonstrate improved calibration where it matters using convnets trained to classify dermoscopy images.