# MaxEnt Loss: Constrained Maximum Entropy for Calibration under Out-of-Distribution Shift

Source: [MaxEnt Loss: Constrained Maximum Entropy for Calibration under Out-of-Distribution Shift](<https://www.sota2.com/research/paper/68fa72711e209131cea565b2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:26:06.010Z

Authors: Dexter Neo, Stefan Winkler, Tsuhan Chen

Published: 2023-10-26T05:10:57.000Z

## Abstract

We present a new loss function that addresses the out-of-distribution (OOD) calibration problem. While many objective functions have been proposed to effectively calibrate models in-distribution, our findings show that they do not always fare well OOD. Based on the Principle of Maximum Entropy, we incorporate helpful statistical constraints observed during training, delivering better model calibration without sacrificing accuracy. We provide theoretical analysis and show empirically that our method works well in practice, achieving state-of-the-art calibration on both synthetic and real-world benchmarks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Classification Calibration on FMoW](<https://www.sota2.com/research/sota/image-classification-calibration-on-fmow.md>): Image Classification Calibration; fMoW; Accuracy: 53.04.

- [Image Classification Calibration on iWildCam](<https://www.sota2.com/research/sota/image-classification-calibration-on-iwildcam.md>): Image Classification Calibration; iWILDCam; Accuracy: 75.11.

- [Image Classification Calibration on CIFAR-10-C](<https://www.sota2.com/research/sota/image-classification-calibration-on-cifar-10-c.md>): Image Classification Calibration; CIFAR-10-C; Accuracy: 71.98.

- [Image Classification Calibration on Tiny ImageNet C](<https://www.sota2.com/research/sota/image-classification-calibration-on-tiny-imagenet-c.md>): Image Classification Calibration; Tiny-ImageNet-C; Accuracy: 21.14.

- [Image Classification Calibration on Camelyon 17](<https://www.sota2.com/research/sota/image-classification-calibration-on-camelyon-17.md>): Image Classification Calibration; CAMELYON 17; Accuracy: 85.67.

- [Image Classification Calibration on CIFAR-100-C](<https://www.sota2.com/research/sota/image-classification-calibration-on-cifar-100-c.md>): Image Classification Calibration; CIFAR-100-C; Accuracy: 48.34.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2310.17159>)

[PDF](<https://arxiv.org/pdf/2310.17159.pdf>)