# Ensemble-Based Dirichlet Modeling for Predictive Uncertainty and Selective Classification

Source: [Ensemble-Based Dirichlet Modeling for Predictive Uncertainty and Selective Classification](<https://www.sota2.com/research/paper/69d5e1427d0172140ec75cfc>)

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

Data updated: 2026-04-09T05:00:13.067Z

Authors: Courtney Franzen, Farhad Pourkamali-Anaraki

Published: 2026-04-07T16:28:20.000Z

## Abstract

Neural network classifiers trained with cross-entropy loss achieve strong predictive accuracy but lack the capability to provide inherent predictive uncertainty estimates, thus requiring external techniques to obtain these estimates. In addition, softmax scores for the true class can vary substantially across independent training runs, which limits the reliability of uncertainty-based decisions in downstream tasks. Evidential Deep Learning aims to address these limitations by producing uncertainty estimates in a single pass, but evidential training is highly sensitive to design choices including loss formulation, prior regularization, and activation functions. Therefore, this work introduces an alternative Dirichlet parameter estimation strategy by applying a method of moments estimator to ensembles of softmax outputs, with an optional maximum-likelihood refinement step. This ensemble-based construction decouples uncertainty estimation from the fragile evidential loss design while also mitigating the variability of single-run cross-entropy training, producing explicit Dirichlet predictive distributions. Across multiple datasets, we show that the improved stability and predictive uncertainty behavior of these ensemble-derived Dirichlet estimates translate into stronger performance in downstream uncertainty-guided applications such as prediction confidence scoring and selective classification.

## 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 on DermaMNIST v2 (test)](<https://www.sota2.com/research/sota/image-classification-on-dermamnist-v2-test.md>): Image Classification; DermaMNIST v2 (test); Accuracy: 79.74.

- [Image Classification on DermaMNIST Retained v2 (test)](<https://www.sota2.com/research/sota/image-classification-on-dermamnist-retained-v2-test.md>): Image Classification; DermaMNIST Retained v2 (test); Accuracy: 89.87.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.06032>)

[PDF](<https://arxiv.org/pdf/2604.06032.pdf>)
