aiqbench analysis
This research introduces a cooperative training framework that successfully disentangles aleatoric and epistemic uncertainties in neural networks, addressing longstanding limitations where standard methods struggle to handle both noise types simultaneously. By combining variance estimation networks with Bayesian inference, the approach improves mean prediction accuracy while remaining robust across diverse datasets without requiring ad-hoc regularization. This advancement offers a practical solution for deploying more reliable AI models in real-world scenarios characterized by imperfect data and model uncertainty.
Cooperative Variance Estimation and Bayesian Neural Networks for Disentangling Aleatoric and Epistemic Uncertainties
arXiv:2505.02743v3 Announce Type: replace Abstract: Real-world data contains aleatoric uncertainty - irreducible noise arising from imperfect measurements or from incomplete knowledge about the data generation process. Mean-variance estimation networks can learn this type of uncertainty but require ad-hoc regularization strategies to avoid overfitting and are unable to predict epistemic uncertainty (model uncertainty). Conversely, Bayesian neural networks predict epistemic uncertainty but are notoriously difficult to train due to the approximate nature of Bayesian inference. We propose to cooperatively train a variance estimation network with a Bayesian neural network and empirically demonstrate that the resulting model disentangles aleatoric and epistemic uncertainties while improving the mean estimation. We demonstrate the effectiveness and scalability of this method across a diverse range of datasets, including a time-dependent heteroscedastic regression dataset we created where the aleatoric uncertainty is known. The proposed method is straightforward to implement, robust, and adaptable to various model architectures. Code is available at https://github.com/bessagroup/VeBNN.
Abstract:Real-world data contains aleatoric uncertainty - irreducible noise arising from imperfect measurements or from incomplete knowledge about the data generation process. Mean-variance estimation networks can learn this type of uncertainty but require ad-hoc regularization strategies to avoid overfitting and are unable to predict epistemic uncertainty (model uncertainty). Conversely, Bayesian neural networks predict epistemic uncertainty but are notoriously difficult to train due to the approximate nature of Bayesian inference. We propose to cooperatively train a variance estimation network with a Bayesian neural network and empirically demonstrate that the resulting model disentangles aleatoric and epistemic uncertainties while improving the mean estimation. We demonstrate the effectiveness and scalability of this method across a diverse range of datasets, including a time-dependent heteroscedastic regression dataset we created where the aleatoric uncertainty is known. The proposed method is straightforward to implement, robust, and adaptable to various model architectures. Code is available at this https URL.
ICML 2026 camera-ready. 38 pages, 26 figures. Code: this https URL
Machine Learning (cs.LG); Machine Learning (stat.ML)
arXiv:2505.02743 [cs.LG]
(or arXiv:2505.02743v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2505.02743 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 255 words
Related coverage
Disentangling Model and Human Data Uncertainty in Apparent Facial Age Estimation
arXiv · Jul 21, 2026
Heteroscedastic Neural Surrogate Modeling for Robust and Rapid Bayesian Inference in Fusion Plasma Diagnostics
arXiv · Aug 21, 2026
Bayesian 3D Steerable CNNs: Enabling Equivariance and Uncertainty Quantification Simultaneously
arXiv · Jun 16, 2026
Uncertainty Estimation using Variance-Gated Distributions
arXiv · Jun 5, 2026