This repo contains a PyTorch implementation of the paper: "Evidential Deep Learning to Quantify Classification Uncertainty"

A curated publication list on evidential deep learning.

Fast and scalable uncertainty quantification for neural molecular property prediction, accelerated optimization, and guided virtual screening.

[ICCV 2021 Oral] Deep Evidential Action Recognition

[ECCV 2022] Dual-Evidential Learning for Weakly-supervised Temporal Action Localization

[ICLR 2024 Spotlight] R-EDL: Relaxing Nonessential Settings of Evidential Deep Learning

Implementation of "Evidential Deep Learning to Quantify Classification Uncertainty" proposing a method to quantify uncertainty in a neural network.

Machine learning models for estimating aleatoric and epistemic uncertainty with evidential and ensemble methods.

[TPAMI 2025] Revisiting Essential and Non-Essential Settings of Evidential Deep Learning

Official implementation of MICCAI2024 paper "Evidential Concept Embedding Models: Towards Reliable Concept Explanations for Skin Disease Diagnosis"

Repository for "Improving evidential deep learning via multi-task learning," published in AAAI2022

Calibrating LLMs with Information-Theoretic Evidential Deep Learning (ICLR 2025)

[ICML 2026] Revisiting Uncertainty: On Evidential Learning for Partially Relevant Video Retrieval

[NeurIPS 2024] Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage?

Official Implementation of Evidential Knowledge Distillation (ICCV 2025)

Evidential Deep Learning Layers for Flux

Hybrid Uncertainty Quantification for Bioactivity Assessment

Our Conflict-aware Evidential Deep Learning (C-EDL) method enhances robustness to OOD and adversarial inputs by combining evidence from metamorphic transformations and reducing evidence when conflicts arise, signalling higher uncertainty.

Official PyTorch implementation of the ECHO Architecture (ADG 2026): Epistemic Cognitive Hierarchical Offloading for Multimodal Industrial Anomaly Detection.

Code for training Mosaico mosquito classifier: fine-tunes a 50M-parameter CNN on a 15-species ISS dataset with modified Evidential Deep Learning for open-set recognition and uncertainty estimation.