evidential deep learning

This work addresses the unreliable and unstable estimation of predictive uncertainty from softmax outputs of neural network classifiers, which adversely affects downstream task performance. The authors propose a novel approach that does not rely on evidential learning losses; instead, it explicitly estimates the parameters of a Dirichlet distribution by aggregating multiple softmax outputs and combining the method of moments with optional maximum likelihood optimization. This formulation effectively decouples uncertainty modeling from classification training. Evaluated across multiple datasets, the method significantly improves the quality of uncertainty estimates and achieves superior performance in tasks such as confidence calibration and selective classification, demonstrating both robustness and practical applicability.

This study systematically evaluates the quality of uncertainty estimation in binary classification models, focusing on two core objectives: probabilistic calibration and out-of-distribution (OOD) detection. Within an approximate Bayesian inference framework, we empirically compare six prominent methods—neural network ensembles, conflicting loss, evidential deep learning, Monte Carlo Dropout, Gaussian process classification, and Dirichlet process mixture models—on controlled synthetic datasets. Results show that all methods achieve strong calibration performance; however, only generative approaches (e.g., Gaussian processes, Dirichlet processes) and explicitly uncertainty-aware models (e.g., evidential deep learning) consistently enhance OOD sensitivity. In contrast, most discriminative deep learning methods exhibit inconsistent OOD responsiveness. To our knowledge, this is the first work to empirically decouple calibration from OOD robustness under a unified benchmark, providing critical empirical evidence and methodological guidance for uncertainty quantification in trustworthy machine learning.

Existing methods struggle to simultaneously achieve high interpretability and accurate uncertainty estimation. This work proposes the first integration of evidential deep learning into Neural Additive Models (NAMs), parameterizing predictive uncertainty via Dirichlet distributions to jointly model aleatoric and epistemic uncertainties in a single forward pass while preserving clear interpretability of feature contributions. The approach naturally supports regression, classification, and generalized additive extensions. Empirical evaluations on both synthetic and real-world datasets demonstrate state-of-the-art predictive performance alongside reliable uncertainty quantification and high model interpretability.

This work addresses the lack of a unified theoretical framework in existing evidential deep learning (EDL) approaches, which obscures their intrinsic connections and design principles. By casting EDL within a generalized Bayesian perspective for the first time, this study elucidates the fundamental nature of distributional uncertainty and systematically dissects the interplay among prior specification, posterior updating, and training objectives. Building on this insight, the authors propose a unified and extensible Generalized Evidential Deep Learning (GEDL) framework. Through component-wise decoupling, GEDL not only integrates and generalizes existing methods but also achieves state-of-the-art performance in classification, uncertainty quantification, and out-of-distribution detection, all while enjoying a rigorous theoretical foundation.

This paper identifies and explains the “cognitive uncertainty collapse” phenomenon—where larger deep learning models exhibit degraded uncertainty quantification despite increased capacity—challenging the prevailing assumption that scale inherently improves uncertainty estimation. Method: The authors first systematically establish implicit ensembling as the primary cause of this collapse; they then propose an explicit multi-layer ensembling framework coupled with submodel decomposition to restore predictive diversity in large vision models (e.g., ViT), thereby recovering calibrated uncertainty estimation. Their approach integrates ViT interpretability analysis, theoretical modeling, and cross-architecture empirical validation (MLP, ResNet, ViT). Contribution/Results: The collapse is consistently reproduced across architectures; the proposed method significantly improves out-of-distribution detection and uncertainty calibration in safety-critical applications, demonstrating robust generalization and advancing principled uncertainty-aware scaling of vision models.

This work addresses the limitation of existing deep learning approaches to uncertainty quantification, which struggle to distinguish between uncertainty arising from missing evidence (vacuity) and that stemming from conflicting evidence (dissonance), while also lacking spatial interpretability. To overcome this, the paper introduces the Uncertainty Activation Map (UAM) framework, which uniquely integrates the concepts of vacuity and dissonance from subjective logic with Full Gradient-based class activation mapping (FullGrad). By leveraging evidential deep learning, UAM generates spatially resolved visualizations of uncertainty that are both theoretically grounded and intuitively interpretable. The method effectively localizes the spatial origins of different uncertainty types across multiple benchmark datasets, offering an explainable visual feedback mechanism for assessing model reliability in complex vision tasks.

This work addresses the challenge of selecting uncertainty representations that align with decision objectives to achieve optimal and trustworthy decisions under state-variable uncertainty. Drawing on decision theory, it systematically analyzes the optimal forms of uncertainty representation for both risk-neutral and risk-averse agents in known and unknown environments, revealing the minimal uncertainty information required under distinct risk preferences. The study innovatively unifies three approaches to epistemic uncertainty—calibrated prediction, confidence-set robust optimization, and Bayesian inference—establishing a theoretical link between uncertainty representation and decision goals. This integration yields a reliable decision-making framework that provides agents with verifiable utility guarantees.

This work addresses the challenge of unreliable predictive uncertainty estimation in deep neural networks, which undermines their trustworthiness in safety-critical applications. The paper presents a systematic survey of uncertainty quantification methods, with a focus on ensemble and approximate Bayesian techniques, and introduces a decoupled “method–metric” framework that unifies the generation of predictive distributions and the aggregation of uncertainties. By integrating diverse approaches—including Bayesian neural networks, Monte Carlo Dropout, deep and efficient ensembles, single-forward methods, evidential networks, conformal prediction, and post-hoc calibration—the study establishes a unified taxonomy and evaluation benchmark. This enables a clear delineation of each method’s theoretical foundations, implementation strategies, empirical performance, and limitations, while also outlining promising directions for uncertainty research in large language models.

This study addresses a critical yet overlooked issue in out-of-distribution (OOD) detection within evidential deep learning (EDL): the sensitivity of uncertainty-mass-based methods to the number of in-distribution (ID) classes. The authors systematically analyze how mismatches in cardinality between ID and OOD label sets introduce evaluation artifacts, particularly in causal language modeling and multiple-choice question answering (MCQA) tasks where ID/OOD boundaries are rigorously defined. They demonstrate for the first time that even a single-class difference can artificially inflate OOD detection performance, yielding AUROC and AUPR biases as high as 0.360 and 0.683, respectively. The work underscores the necessity of strictly controlling class cardinality to ensure reliable OOD evaluation in EDL frameworks and validates these findings across EDL, IB-EDL, and Dirichlet belief models on standard MCQA benchmarks.

Automated mental health prediction using textual data has shown promising results with deep learning and large language models. However, deploying these models in high-stakes real-world settings remains challenging, as existing approaches largely rely on semantic representations and often produce overconfident predictions under ambiguous, noisy, or shifted data. Moreover, most methods lack reliable uncertainty estimation, undermining trust in risk-sensitive mental health applications. To address these limitations, we formulate the task as a multi-view learning problem that integrates semantic information from encoder-only models with higher-level reasoning information from decoder-only models, where reasoning-aware representations and uncertainty modeling are obtained in a trustworthy manner. To ensure reliable fusion, we adopt an evidential learning framework based on Subjective Logic to explicitly model uncertainty and introduce an evidential fusion strategy that balances complementary views while discounting unreliable evidence. Benchmarking on three real-world datasets, Dreaddit, SDCNL, and DepSeverity, reports accuracies of 0.835, 0.731, and 0.751, respectively, demonstrating its potential for reliable mental health prediction. Additional experiments on robustness to noise and case studies for interpretability confirm that our proposed framework not only improves predictive performance but also provides trustworthy uncertainty estimates and human-understandable reasoning signals, making it suitable for risk-sensitive applications in mental health assessment.