Clustering Methods for ANN （Ann 中的聚合方法）

金戈铁马嘟

于 2023-02-27 17:45:30 发布

阅读量123

点赞数

分类专栏： ANN 文章标签：机器学习深度学习人工智能

本文链接：https://blog.csdn.net/m0_66436090/article/details/129246419

版权

文章介绍了在大规模近似最近邻搜索中，如何使用聚类技术加速找到查询点的最近邻。通过k-means、层次聚类、DBSCAN等传统方法以及基于深度学习的DeepEmbeddedClustering、VAE聚类等方法，提高搜索效率并减少计算复杂性。

摘要由CSDN通过智能技术生成

简介(Introduction)

Clustering is an important technique used in large-scale approximate nearest neighbor search to accelerate the process of finding the closest neighbors of a query point within a large dataset.

In approximate nearest neighbor search, the goal is to find the k-nearest neighbors of a query point, where k is a small integer. The brute force approach involves computing the distance between the query point and each data point in the dataset, which can be computationally expensive for large datasets.

Clustering helps to reduce the computational complexity of the search by dividing the dataset into smaller groups or clusters, based on their similarity. This is typically done using clustering algorithms like k-means or hierarchical clustering.

Once the dataset has been clustered, the nearest cluster(s) to the query point can be identified efficiently, and the search can be restricted to the data points within those clusters. This red