The paper examines the common practice of aggregating multiple labels per instance into a single ‘true’ label for supervised learning. By creating a theoretical model, the authors show that using the full, non‑aggregated label information can make it easier to train well‑calibrated models, though the benefits depend on the specific problem. They predict when non‑aggregated labels will improve learning and validate these predictions on real datasets.
arXiv:2512.12870v2 Announce Type: replace-cross Abstract: Active Learning (AL) is commonly used in applications where labeling data is expensive or time-consuming. In practice, however, labels are of...
arXiv:2510.16211v2 Announce Type: replace Abstract: Label noise is a common problem in real-world datasets, affecting both model training and validation. Clean data are essential for achieving strong...
arXiv:2509. 05130v2 Announce Type: replace Abstract: In classification problems, models are trained to predict a class label based on the input data features.
arXiv:2505. 20178v2 Announce Type: replace-cross Abstract: Prediction-Powered Inference (PPI) is a popular strategy for combining gold-standard and possibly noisy pseudo-labels to perform statistical estimation.
FedCC is a new algorithm for distillation-based federated learning that tackles label distribution skew by allowing clients to mark ambiguous samples as 'unknown' instead of forcing a potentially wrong classification. By adding this extra class and calibrating pseudo-labels on a public dataset, FedCC balances confidence across majority and minority classes. Experiments show that FedCC outperforms existing methods, achieving 67.3% accuracy even when each client has data from only one of ten classes, whereas baselines drop to near-random performance.
arXiv:2607. 11541v1 Announce Type: new Abstract: We introduce a straightforward yet effective method to empirically study memorization in deep neural networks for classification tasks.
arXiv:2609.26631v1 Announce Type: new Abstract: Accurate ground-based cloud classification is important for atmospheric monitoring, solar-energy forecasting, aviation weather assessment, and climate...
arXiv:2606. 11699v1 Announce Type: new Abstract: The performance of machine learning and deep learning models largely depends on the quality of the training data.
Federated Learning (FL) enables distributed clients to collaboratively train models without sharing raw data, making it promising for leveraging massive devices in communication networks. In distillat...
The paper investigates the impact of label‑flipping attacks on distributed machine learning, where an adversary can only flip a limited number of training labels. It formalizes the attack as a per‑round constrained optimization problem, derives a greedy label‑selection rule for logistic regression, and shows that this rule is provably optimal under mean aggregation. Experiments demonstrate that optimized label flipping can significantly degrade model accuracy, outperforming random flips, and that the attack transfers to other robust aggregators such as coordinate‑wise median and trimmed mean.
arXiv:2511. 14117v2 Announce Type: replace Abstract: Supervised classifiers output a distribution over classes but are typically trained against a single label obtained by collapsing multiple annotators into a majority vote.
arXiv:2606. 14965v1 Announce Type: new Abstract: Synthetic instance-dependent label noise (IDN) benchmarks are widely used to evaluate noisy-label learning methods, yet existing approaches typically generate noise through imperfect annotators or classifier raters, leaving the source of ambiguity implicit.