Score-based divergences have been widely used in machine learning and statistics applications. Despite their empirical success, a blindness problem has been observed when using these for multi-modal distributions. In this work, we discuss the blindness problem and propose a new family of divergences that can mitigate the blindness problem. We illustrate our proposed divergence in the context of density estimation and report improved performance compared to traditional approaches.

翻译：基于评分的散度在机器学习和统计学应用中已被广泛使用。尽管其经验上取得了成功，但在处理多模态分布时，这些方法存在一个被称为“盲区”的问题。本研究探讨了该盲区问题，并提出了一类新的散度族，能够有效缓解盲区现象。我们在密度估计的背景下展示了所提出的散度方法，并报告了相较于传统方法性能提升的结果。