![[2510.18322] Uncertainty Estimation by Flexible Evidential Deep Learning](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2510.18322] Uncertainty Estimation by Flexible Evidential Deep Learning
Summary
This paper introduces Flexible Evidential Deep Learning (F-EDL), enhancing uncertainty quantification in machine learning by extending the Dirichlet distribution model for improved robustness and generalization in complex scenarios.
Why It Matters
Uncertainty quantification is vital in high-stakes machine learning applications, where inaccurate predictions can have serious consequences. F-EDL addresses limitations of traditional evidential deep learning methods, offering a more adaptable approach to uncertainty modeling, which is crucial for developing reliable AI systems.
Key Takeaways
- F-EDL generalizes the Dirichlet distribution for better uncertainty modeling.
- The method enhances performance in diverse evaluation settings, including noisy and long-tailed data.
- Improved uncertainty quantification can lead to safer AI applications in critical areas.
Computer Science > Machine Learning arXiv:2510.18322 (cs) [Submitted on 21 Oct 2025 (v1), last revised 20 Feb 2026 (this version, v2)] Title:Uncertainty Estimation by Flexible Evidential Deep Learning Authors:Taeseong Yoon, Heeyoung Kim View a PDF of the paper titled Uncertainty Estimation by Flexible Evidential Deep Learning, by Taeseong Yoon and 1 other authors View PDF HTML (experimental) Abstract:Uncertainty quantification (UQ) is crucial for deploying machine learning models in high-stakes applications, where overconfident predictions can lead to serious consequences. An effective UQ method must balance computational efficiency with the ability to generalize across diverse scenarios. Evidential deep learning (EDL) achieves efficiency by modeling uncertainty through the prediction of a Dirichlet distribution over class probabilities. However, the restrictive assumption of Dirichlet-distributed class probabilities limits EDL's robustness, particularly in complex or unforeseen situations. To address this, we propose \textit{flexible evidential deep learning} ($\mathcal{F}$-EDL), which extends EDL by predicting a flexible Dirichlet distribution -- a generalization of the Dirichlet distribution -- over class probabilities. This approach provides a more expressive and adaptive representation of uncertainty, significantly enhancing UQ generalization and reliability under challenging scenarios. We theoretically establish several advantages of $\mathcal{F}$-EDL and empirically...
