Tag
该论文研究了不同训练策略与损失函数对长尾/不平衡数据上集值分类器（SVC）性能的影响，并改进了 odd-KK-out（OKO）损失，将其计算开销从3.2×降至1.36×，显著提升了多个SVC方法的宏平均准确率。
The paper proposes a method for maximizing the partial area under the ROC curve (pAUC) using only positive and unlabeled data, deriving an empirical pAUC estimator from positive and marginal densities within an empirical risk minimization framework, and demonstrates its effectiveness on ten real-world datasets.
Barge introduces a single-stage objective combining bounded adjustment and reliability-guided embeddings to address imbalanced learning with noisy labels, achieving competitive performance on benchmarks like CIFAR-10 and Tiny ImageNet.