Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning

• URL: http://arxiv.org/abs/2602.01477v1
• Date: Sun, 01 Feb 2026 22:57:39 GMT
• Title: Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning
• Authors: Pietro Carlotti, Nevena Gligić, Arya Farahi,
• Abstract summary: Evidential Deep Learning (EDL) is a popular framework for uncertainty-aware classification that models predictive uncertainty via Dirichlet distributions parameterized by neural networks.<n>We show that EDL training corresponds to amortized variational inference in a hierarchical Bayesian model with a tempered pseudo-likelihood.<n>We introduce Density-Informed Pseudo-count EDL (DIP-EDL), a new parametrization that decouples class prediction from the magnitude of uncertainty.
• Score: 1.9435397960631862
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Evidential Deep Learning (EDL) is a popular framework for uncertainty-aware classification that models predictive uncertainty via Dirichlet distributions parameterized by neural networks. Despite its popularity, its theoretical foundations and behavior under distributional shift remain poorly understood. In this work, we provide a principled statistical interpretation by proving that EDL training corresponds to amortized variational inference in a hierarchical Bayesian model with a tempered pseudo-likelihood. This perspective reveals a major drawback: standard EDL conflates epistemic and aleatoric uncertainty, leading to systematic overconfidence on out-of-distribution (OOD) inputs. To address this, we introduce Density-Informed Pseudo-count EDL (DIP-EDL), a new parametrization that decouples class prediction from the magnitude of uncertainty by separately estimating the conditional label distribution and the marginal covariate density. This separation preserves evidence in high-density regions while shrinking predictions toward a uniform prior for OOD data. Theoretically, we prove that DIP-EDL achieves asymptotic concentration. Empirically, we show that our method enhances interpretability and improves robustness and uncertainty calibration under distributional shift.

Related papers

• Epistemic Uncertainty Quantification for Pre-trained VLMs via Riemannian Flow Matching [3.0708725114491293]
  Vision-Language Models (VLMs) are typically deterministic in nature and lack intrinsic mechanisms to quantify uncertainty.<n>We theoretically motivate negative log-density of an embedding as a proxy for the epistemic uncertainty, where low-density regions signify model ignorance.<n>We empirically demonstrate that REPVLM achieves near-perfect correlation between uncertainty and prediction error, significantly outperforming existing baselines.
  arXiv  Detail & Related papers  (2026-01-29T12:58:42Z)
• Bridging the Gap Between Bayesian Deep Learning and Ensemble Weather Forecasts [100.26854618129039]
  Weather forecasting is fundamentally challenged by the chaotic nature of the atmosphere.<n>Recent advances in Bayesian Deep Learning (BDL) offer a promising but often disconnected alternative.<n>We bridge these paradigms through a unified hybrid BDL framework for ensemble weather forecasting.
  arXiv  Detail & Related papers  (2025-11-18T07:49:52Z)
• Revisiting Essential and Nonessential Settings of Evidential Deep Learning [70.82728812001807]
  Evidential Deep Learning (EDL) is an emerging method for uncertainty estimation. We propose Re-EDL, a simplified yet more effective variant of EDL.
  arXiv  Detail & Related papers  (2024-10-01T04:27:07Z)
• Inflationary Flows: Calibrated Bayesian Inference with Diffusion-Based Models [0.0]
  We show how diffusion-based models can be repurposed for performing principled, identifiable Bayesian inference.<n>We show how such maps can be learned via standard DBM training using a novel noise schedule.<n>The result is a class of highly expressive generative models, uniquely defined on a low-dimensional latent space.
  arXiv  Detail & Related papers  (2024-07-11T19:58:19Z)
• Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage? [35.15844215216846]
  EDL methods are trained to learn a meta distribution over the predictive distribution by minimizing a specific objective function. Recent studies identify limitations of the existing methods to conclude their learned uncertainties are unreliable. We provide a sharper understanding of the behavior of a wide class of EDL methods by unifying various objective functions. We conclude that even when EDL methods are empirically effective on downstream tasks, this occurs despite their poor uncertainty quantification capabilities.
  arXiv  Detail & Related papers  (2024-02-09T03:23:39Z)
• Knowledge from Uncertainty in Evidential Deep Learning [10.751990848772028]
  This work reveals an evidential signal that emerges from the uncertainty value in Evidential Deep Learning (EDL) EDL is one example of a class of uncertainty-aware deep learning approaches designed to provide confidence about the current test sample.
  arXiv  Detail & Related papers  (2023-10-19T11:41:52Z)
• Excess risk analysis for epistemic uncertainty with application to variational inference [110.4676591819618]
  We present a novel EU analysis in the frequentist setting, where data is generated from an unknown distribution. We show a relation between the generalization ability and the widely used EU measurements, such as the variance and entropy of the predictive distribution. We propose new variational inference that directly controls the prediction and EU evaluation performances based on the PAC-Bayesian theory.
  arXiv  Detail & Related papers  (2022-06-02T12:12:24Z)
• DEUP: Direct Epistemic Uncertainty Prediction [56.087230230128185]
  Epistemic uncertainty is part of out-of-sample prediction error due to the lack of knowledge of the learner. We propose a principled approach for directly estimating epistemic uncertainty by learning to predict generalization error and subtracting an estimate of aleatoric uncertainty.
  arXiv  Detail & Related papers  (2021-02-16T23:50:35Z)
• The Hidden Uncertainty in a Neural Networks Activations [105.4223982696279]
  The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution correlates with a model's epistemic uncertainty, thus indicating its ability to generalise to novel inputs.
  arXiv  Detail & Related papers  (2020-12-05T17:30:35Z)
• Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift [100.52588638477862]
  We develop an approximate Bayesian inference scheme based on posterior regularisation. We demonstrate the utility of our method in the context of transferring prognostic models of prostate cancer across globally diverse populations.
  arXiv  Detail & Related papers  (2020-06-26T13:50:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.