H$^2$EDL: Hyper Evidential Deep Learning for Hierarchical Classification

• URL: http://arxiv.org/abs/2608.18185v1
• Date: Tue, 18 Aug 2026 08:42:32 GMT
• Title: H$^2$EDL: Hyper Evidential Deep Learning for Hierarchical Classification
• Abstract summary: A new uncertainty representation, H$2$EDL, can capture both fine-grained classes and intermediate concepts.<n>On FGVC-Aircraft and DERM12345, H$2$EDL reduces calibration error by approximately half compared with cross-entropy baselines.
• Score: 19.611209640933435
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Fine-grained recognition often involves hierarchical label spaces, where a model may be confident about a coarse semantic concept while remaining uncertain among its descendant classes. Such structured ambiguity requires uncertainty representations that capture both fine-grained classes and intermediate concepts. However, existing tools each capture only half of it: flat evidential classifiers quantify total ignorance with a single vacuity on the leaf frame, and hierarchical classifiers propagate point probabilities with no notion of evidence. Hyper-opinions would unify the two, but their general form is exponential in the label count, and existing hyper-evidential networks either require composite labels to be supplied in the training data or read them off an unstructured weight pattern, with no principled notion of which composites deserve mass. We observe that the taxonomy itself is the missing hyperdomain. Its subtrees and leaf singletons form a linear-size focal family, and one local Dirichlet opinion per branching node induces every composite mass in closed form. The resulting model, H$^2$EDL, can be interpreted in two complementary ways using the same set of parameters. From a prediction perspective, it functions as a hierarchical classifier that preserves consistency across different levels of the label tree. From a probabilistic perspective, it defines a valid tree-structured hyper-opinion, where the mass assigned to each node represents the belief that reaches that node but does not provide sufficient confidence to further specialize into its descendants. On FGVC-Aircraft and DERM12345, H$^2$EDL reduces calibration error by approximately half compared with cross-entropy baselines, with the improvement becoming more pronounced at deeper hierarchy levels and under larger training budgets.

Related papers

• Diffusion Models and Concept Formation [2.9491967009603255]
  We argue that diffusion models implicitly perform the same computation as Cobweb.<n>We find that Cobweb learns a discrete tree incrementally, whereas a diffusion model encodes a continuous, interpolable hierarchy in a single learned score field.<n>This reframes diffusion as a cognitive model of concept formation and offers Cobweb a continuous, scalable instantiation.
  arXiv  Detail & Related papers  (2026-09-11T16:43:48Z)
• Beyond Flat Labels: Level-Restricted Contrastive Learning for Hierarchical Fine-Grained Vision Classification [50.20892407648858]
  Multimodal contrastive learning has enabled zero-shot visual classification by aligning images with textual categories.<n>Existing methods often produce predictions that are inconsistent across taxonomic levels.<n>We propose a group-balanced design, ensuring each taxonomic level receives adequate optimization.<n>We train our model with TreeOfLife-10M based on BioCLIP and evaluate it across multiple hierarchical classification benchmarks.
  arXiv  Detail & Related papers  (2026-06-20T02:12:05Z)
• Learn from your own latents and not from tokens: A sample-complexity theory [53.5821824211418]
  We study networks trained to predict their emphown latent representations of related views or masked regions.<n>We show that latent prediction achieves this with a number of samples emphconstant in $L$, up to logarithmic factors.<n>This suggests that explicit stacking such as H-JEPA is largely redundant.
  arXiv  Detail & Related papers  (2026-05-26T22:16:42Z)
• Hierarchically Robust Zero-shot Vision-language Models [74.79155369056639]
  Vision-Language Models (VLMs) can perform zero-shot classification but are susceptible to adversarial attacks.<n>Existing approaches align fixed text embeddings with an image embedding, sacrificing natural performance and robustness.<n>We propose a novel adversarial fine-tuning framework based on hierarchical embeddings and adversarially robust alignment of image-text modalities.
  arXiv  Detail & Related papers  (2026-04-20T21:42:14Z)
• Semantic Level of Detail: Multi-Scale Knowledge Representation via Heat Kernel Diffusion on Hyperbolic Manifolds [0.0]
  We introduce Semantic Level of Detail (SLoD), a framework that defines a continuous zoom operator via heat kernel diffusion on the Poincaré ball.<n>We show that spectral gaps in the graph Laplacian induce emergent scale boundaries.<n>On synthetic hierarchies, our boundary scanner recovers planted levels with ARI up to 1.00, with detection degrading gracefully near the information-theoretic Kesten-Stigum threshold.
  arXiv  Detail & Related papers  (2026-03-09T21:54:08Z)
• Learning Consistent Taxonomic Classification through Hierarchical Reasoning [61.372270953201955]
  We propose a two-stage, hierarchy-based reasoning framework designed to improve leaf-level accuracy and hierarchical consistency in taxonomic classification.<n>Our framework, implemented on the Qwen2.5-VL-7B model, outperforms its original 72B counterpart by over 10% in both leaf-level and hierarchical consistency accuracy.
  arXiv  Detail & Related papers  (2026-01-21T03:00:00Z)
• Climbing the label tree: Hierarchy-preserving contrastive learning for medical imaging [0.0]
  We present a hierarchy-preserving contrastive framework that makes the label tree a first-class training signal and an evaluation target.<n>Our approach introduces two plug-in objectives: Hierarchy-Weighted Contrastive (HWC), which scales positive/negative pair strengths by shared ancestors, and Level-Aware Margin (LAM), a prototype margin that separates ancestor groups across levels.
  arXiv  Detail & Related papers  (2025-11-05T18:53:25Z)
• Visually Consistent Hierarchical Image Classification [37.80849457554078]
  Hierarchical classification predicts labels across multiple levels of a taxonomy, e.g., from coarse-level 'Bird' to mid-level 'Hummingbird' to fine-level 'Green hermit'
  arXiv  Detail & Related papers  (2024-06-17T14:56:51Z)
• Information-Theoretic Limits and Strong Consistency on Binary Non-uniform Hypergraph Stochastic Block Models [0.0]
  unsupervised classification problem in random hypergraphs under the non-uniform Hypergraph the Block Model (HSBM)<n>In this paper, the information-theoretic limits on the clustering accuracy and the strong consistency threshold are established.
  arXiv  Detail & Related papers  (2023-06-12T03:38:25Z)
• Label Relation Graphs Enhanced Hierarchical Residual Network for Hierarchical Multi-Granularity Classification [10.449261628173229]
  We study the HMC problem in which objects are labeled at any level of the hierarchy. We propose a hierarchical residual network (HRN) in which residual connections are added to features of children levels.
  arXiv  Detail & Related papers  (2022-01-10T07:17:24Z)
• Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones [64.75766944882389]
  We present ConE (Cone Embedding), a KG embedding model that is able to simultaneously model multiple hierarchical as well as non-hierarchical relations in a knowledge graph. In particular, ConE uses cone containment constraints in different subspaces of the hyperbolic embedding space to capture multiple heterogeneous hierarchies. Our approach yields new state-of-the-art Hits@1 of 45.3% on WN18RR and 16.1% on DDB14 (0.231 MRR)
  arXiv  Detail & Related papers  (2021-10-28T07:16:08Z)
• Hierarchical nucleation in deep neural networks [67.85373725288136]
  We study the evolution of the probability density of the ImageNet dataset across the hidden layers in some state-of-the-art DCNs. We find that the initial layers generate a unimodal probability density getting rid of any structure irrelevant for classification. In subsequent layers density peaks arise in a hierarchical fashion that mirrors the semantic hierarchy of the concepts.
  arXiv  Detail & Related papers  (2020-07-07T14:42:18Z)

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.