mlsanityData to 7 Oct 2026

Diverse Ensembles Improve Calibration

arXiv 2007.04206 ↗First posted 8 Jul 2020cs.LG stat.MLcited by 23

Not on the map. Why?

Abstract (arXiv)

Modern deep neural networks can produce badly calibrated predictions, especially when train and test distributions are mismatched. Training an ensemble of models and averaging their predictions can help alleviate these issues. We propose a simple technique to improve calibration, using a different data augmentation for each ensemble member. We additionally use the idea of `mixing' un-augmented and augmented inputs to improve calibration when test and training distributions are the same. These simple techniques improve calibration and accuracy over strong baselines on the CIFAR10 and CIFAR100 benchmarks, and out-of-domain data from their corrupted versions.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). First posted on arXiv 8 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 9; 2022: 4; 2023: 6; 2024: 1; 2025: 3.

23 papers.

Sort:

  1. 2021Robust fine-tuning of zero-shot modelscited by 785
  2. 2023Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questionscited by 249
  3. 2022Agree to Disagree: Diversity through Disagreement for Better Transferabilitycited by 69
  4. 2021DeepGaze IIE: Calibrated prediction in and out-of-domain for state-of-the-art saliency modelingcited by 55
  5. 2021DICE: Diversity in Deep Ensembles via Conditional Redundancy Adversarial Estimationcited by 48
  6. 2021Knowing More About Questions Can Help: Improving Calibration in Question Answeringcited by 47
  7. 2023Weighted Ensemble Models Are Strong Continual Learnerscited by 41
  8. 2022Calibrated ensembles can mitigate accuracy tradeoffs under distribution shiftcited by 34
  9. 2021Orthogonal Ensemble Networks for Biomedical Image Segmentationcited by 16
  10. 2021Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentationcited by 13
  11. 2021Combining Diverse Feature Priorscited by 12
  12. 2025Query-Level Uncertainty in Large Language Modelscited by 12
11 more
  1. 2021Should Ensemble Members Be Calibrated?cited by 11
  2. 2023FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensemblingcited by 11
  3. 2022Improving robustness and calibration in ensembles with diversity regularizationcited by 8
  4. 2024DIVE: Subgraph Disagreement for Graph Out-of-Distribution Generalizationcited by 7
  5. 2021Combining Different V1 Brain Model Variants to Improve Robustness to Image Corruptions in CNNscited by 5
  6. 2022Bayesian posterior approximation with stochastic ensemblescited by 5
  7. 2023Multi-Symmetry Ensembles: Improving Diversity and Generalization via Opposing Symmetriescited by 1
  8. 2023Diversifying Deep Ensembles: A Saliency Map Approach for Enhanced OOD Detection, Calibration, and Accuracycited by 1
  9. 2025Parameter-Efficient Construction of the Rashomon Slice for Concept Bottleneck Modelscited by 1
  10. 2023Source-free Domain Adaptation Requires Penalized Diversitycited by 0
  11. 2025WWAggr: A Window Wasserstein-based Aggregation for Ensemble Change Point Detectioncited by 0