Article

Quantifying Geometric Transferability: An Equivariant Diagnostic Framework with Application to Waste Classification

2026-03-31

Abstract excerpt

<title>Abstract</title> <p> Deep neural networks often experience severe performance degradation when deployed across domains with different acquisition conditions, a phenomenon known as domain shift. In real-world waste classification, models trained on clean, laboratory-controlled datasets frequently fail when transferred to cluttered landfill environments, and vice versa. Existing explanations based on statis...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: a9bd87cd-2904-50d8-9f61-e7cc71971a1d
DOI: 10.21203/rs.3.rs-9179600/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Quantifying Geometric Transferability: An Equivariant Diagnostic Framework with Application to Waste ClassificationDOI 10.21203/rs.3.rs-9179600/v1Read

Geographic Bias Analysis and Cross-Domain Generalization in Deep Learning-Based Building Damage AssessmentSemantic proximity Read

Machine-Learning-Driven Formation Damage Classification Based on SEM Images and Petrophysical DataSemantic proximity Read

Instructional Mask AutoEncoder: A Powerful Pretrained Model for Hyperspectral Image ClassificationSemantic proximity Read

Scaffolded representation learning in deep networksSemantic proximity Read

Navigating the Complexity Trap: Deep Learning Optimization for Mapping Historic Surface Mines in Appalachian TerrainSemantic proximity Read

Towards CNN Representations for Small Mass Spectrometry Data Classification: From Transfer Learning to Cumulative LearningSemantic proximity Read

Specxai - Spectral Interpretability of Deep Learning ModelsSemantic proximity Read

A novel meta-learning method based on relation network for train bearing fault diagnosisSemantic proximity Read

Large Language Models as Materials Science Adapted LearnersSemantic proximity Read

Achieving Explainable, Scalable, and Robust Machine Learning for Real-World ApplicationsSemantic proximity Read

Benchmarking and Stress-Testing Off-the-Shelf Vision Embeddings for Surface Re-Identification in Open-Source IntelligenceSemantic proximity Read