用双峰高斯分布做软标签，提升分类模型鲁棒性。
Distributional Loss for Robust Classification
- 将分类输出建模为双峰高斯分布，替代单一标签
- 低数据场景下显著提升模型鲁棒性，优于标准损失
- 无需额外标签信息，兼容主流训练流程
本文提出一种面向监督分类任务的新损失范式。不同于传统方法直接将每个样本映射到单一标签，该方法在所有分类器输出上定义一个双峰高斯分布作为优化目标。这种更柔和的标签形式隐式捕捉类别模糊性，缓解过拟合，并促进学习更鲁棒的决策边界，且无需额外标签信息。实验表明，在各类设置中均实现一致的鲁棒性提升，尤其在低数据场景下效果显著，同时仅需对标准训练流程进行最小修改。
原文摘要 · Abstract (English)
This paper proposes a novel loss concept for supervised classification tasks. Rather than enforcing a direct mapping from each input sample to a single assigned label, we define an optimization objective over all classifier outputs as a bimodal Gaussian distribution. This softer target formulation implicitly captures class ambiguity, mitigates overfitting, and encourages the learning of more robust decision boundaries, all without requiring additional label information. Experimental results demonstrate consistent improvements in robustness, with particularly pronounced gains in low-data regimes, while requiring only minimal modifications to standard training pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。