发现密度聚类中聚类数与半径关系呈单峰分布，提升参数调优效率
Unimodal Strategies in Density-Based Clustering
- 利用三元搜索法高效寻找最优邻域半径
- 在高维大规模数据上验证了方法的稳定性和有效性
- 适合需要自动调参的高维聚类任务，如文本、音频、视觉
密度聚类方法在处理含噪声或任意分布的真实数据时，通常优于基于质心的方法。本研究揭示了密度聚类中聚类数与核心点邻域半径之间存在近似单峰关系，并在特定条件下给出理论支持。基于此性质，我们提出一种基于三元搜索的新策略，更高效地确定半径值，尤其适用于高维、大规模数据的参数调优。我们在涵盖自然语言处理、音频和计算机视觉的多种大规模高维任务中进行了广泛验证，结果表明该方法具有良好的实用性和鲁棒性。该工作不仅提升了密度聚类的参数控制能力，也深化了对聚类参数间关系的理解。代码已开源：https://github.com/oronnir/UnimodalStrategies。
原文摘要 · Abstract (English)
Density-based clustering methods often surpass centroid-based counterparts, when addressing data with noise or arbitrary data distributions common in real-world problems. In this study, we reveal a key property intrinsic to density-based clustering methods regarding the relation between the number of clusters and the neighborhood radius of core points - we empirically show that it is nearly unimodal, and support this claim theoretically in a specific setting. We leverage this property to devise new strategies for finding appropriate values for the radius more efficiently based on the Ternary Search algorithm. This is especially important for large scale data that is high-dimensional, where parameter tuning is computationally intensive. We validate our methodology through extensive applications across a range of high-dimensional, large-scale NLP, Audio, and Computer Vision tasks, demonstrating its practical effectiveness and robustness. This work not only offers a significant advancement in parameter control for density-based clustering but also broadens the understanding regarding the relations between their guiding parameters. Our code is available at https://github.com/oronnir/UnimodalStrategies.
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。