arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.05171cs.LGcs.AIstat.ML
收藏

快速且急切的 k-Medoids 聚类:PAM、CLARA 和 CLARANS 算法的 O(k) 运行时间改进

Fast and Eager k-Medoids Clustering: O(k) Runtime Improvement of the PAM, CLARA, and CLARANS Algorithms

TU Dortmund University(多特蒙德工业大学);KU Leuven(鲁汶大学)

Erich Schubert, Peter J. Rousseeuw

AI总结:

本文通过改进 PAM 算法的 SWAP 阶段实现 O(k) 倍加速,并提出急切交换策略及更快初始化方法,同时惠及 CLARA 和 CLARANS,在 k=100、200 时分别获得 458 倍和 1191 倍加速。

摘要:

对非欧几里得数据进行聚类是困难的,除了层次聚类之外,最常用的算法之一是流行的 Partitioning Around Medoids (PAM),也简称为 k-medoids 聚类。在欧几里得几何中,k-means 所使用的均值是簇中心的良好估计量,但对于任意不相似度而言,这种均值并不存在。PAM 改用 medoid,即簇中与所有其他对象具有最小不相似度的对象。这种中心性概念可以用于任意(不)相似度,因此对许多领域和应用具有高度相关性。PAM 的一个关键问题在于其高昂的运行时间成本。我们提出对 PAM 算法的修改,在算法的第二阶段(“SWAP”阶段)实现 O(k) 倍的加速,同时仍能得到与原始 PAM 算法相同的结果。如果我们放宽对所执行交换的选择(同时保持可比较的质量),我们可以通过在每次迭代中急切地执行额外交换来进一步加速算法。借助显著加快的 SWAP,我们现在可以探索更快的初始化策略,因为(i)经典的(“BUILD”)初始化现在成为瓶颈,并且(ii)我们的交换足够快,足以弥补较差的起始条件。我们还展示了 CLARA 和 CLARANS 算法如何从所提出的修改中受益。虽然我们在本工作中没有研究该方法的并行化,但它可以很容易地与先前在大数据上使用 PAM 和 CLARA 的方法相结合(其中一些方法将 PAM 作为子程序使用,因此可以立即从这些改进中受益),在这些场景中,高 k 值下的性能变得越来越重要。在 k=100,200 的真实数据实验中,与原始 PAM SWAP 算法相比,我们观察到分别有 458 倍和 1191 倍的加速,这使得 PAM 能够适用于更大的数据集,尤其是更高的 k 值。

英文摘要:

Clustering non-Euclidean data is difficult, and one of the most used algorithms besides hierarchical clustering is the popular algorithm Partitioning Around Medoids (PAM), also simply referred to as k-medoids clustering. In Euclidean geometry the mean-as used in k-means-is a good estimator for the cluster center, but this does not exist for arbitrary dissimilarities. PAM uses the medoid instead, the object with the smallest dissimilarity to all others in the cluster. This notion of centrality can be used with any (dis-)similarity, and thus is of high relevance to many domains and applications. A key issue with PAM is its high run time cost. We propose modifications to the PAM algorithm that achieve an O(k)-fold speedup in the second ("SWAP") phase of the algorithm, but will still find the same results as the original PAM algorithm. If we relax the choice of swaps performed (while retaining comparable quality), we can further accelerate the algorithm by eagerly performing additional swaps in each iteration. With the substantially faster SWAP, we can now explore faster initialization strategies, because (i) the classic ("BUILD") initialization now becomes the bottleneck, and (ii) our swap is fast enough to compensate for worse starting conditions. We also show how the CLARA and CLARANS algorithms benefit from the proposed modifications. While we do not study the parallelization of our approach in this work, it can easily be combined with earlier approaches to use PAM and CLARA on big data (some of which use PAM as a subroutine, hence can immediately benefit from these improvements), where the performance with high k becomes increasingly important. In experiments on real data with k=100,200, we observed a 458x respectively 1191x speedup compared to the original PAM SWAP algorithm, making PAM applicable to larger data sets, and in particular to higher k.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑