arXiv Machine Learning
Neural Collapse on Imbalanced Datasets and Open-Set Scenario
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
1d ago
Comparing Classification and Feature Learning by Analyzing Their Loss Functions in Normalized Spaces with Different Scale Factors
arXiv:2611.07741v1 Announce Type: new Abstract: Classification and feature learning are two closely interrelated fundamental tasks. Feature learning in supervised settings is typically achieved by tr...
arXiv Machine Learning
Oct 21
Using Set Learning to Improve Set-Valued Classifiers on Long-tailed Data
arXiv:2610.27443v1 Announce Type: new Abstract: Set-valued classification (SVC) offers a practical approach to communicating uncertainty while maintaining accuracy by predicting variable-sized sets p...
arXiv AI
Jun 16
Beyond Rebalancing: Benchmarking Binary Classifiers Under Class Imbalance Without Rebalancing Techniques
arXiv:2509. 07605v2 Announce Type: replace-cross Abstract: Class imbalance poses a significant challenge to supervised classification, particularly in critical domains like medical diagnostics and anomaly detection where minority class instances are rare.
arXiv Computer Vision
Oct 21
DMDSC: A Dynamic-Margin Deep Simplex Classifier for Open-Set Recognition on Medical Image Datasets
arXiv:2605.00675v3 Announce Type: replace Abstract: Medical imaging datasets are often characterized by extreme class imbalances, where rare pathologies are significantly underrepresented compared to...
arXiv Statistics ML
Oct 1
Constrained Classification and Policy Learning
The paper investigates the consistency of surrogate loss methods for classification and policy learning when the set of admissible classifiers is constrained, such as by interpretability or fairness requirements. It shows that hinge loss is the only surrogate that preserves consistency when constraints limit only the prediction set, but consistency can fail if constraints also restrict the functional form. The authors derive conditions guaranteeing consistency for hinge-risk-minimizing classifiers and use these results to design efficient hinge-loss-based procedures for monotone classification problems.
arXiv Machine Learning
Sep 22
Rethinking Class Imbalance for Single-Cell Foundation Models: A Systematic Benchmark Across Architectures and Long-Tail Loss Functions
The paper benchmarks six long‑tail loss functions—cross‑entropy, weighted CE, class‑balanced loss, focal loss, LDAM, and logit‑adjusted softmax—across three single‑cell foundation model architectures (scGPT, scBERT, Geneformer) and three datasets (Multiple Sclerosis, Zheng68K, human Pancreas). It shows that overall accuracy masks systematic failures on rare, disease‑relevant cell types, with a consistent gap between overall accuracy, Macro‑F1, and rare‑class recall under plain cross‑entropy. The study identifies two distinct regimes of rare‑class failure, predicts reweighting efficacy by absolute training‑set size, and finds class‑balanced loss and LDAM to be the most reliable across all settings.