Summary of Kernel Kmeans Clustering Splits For End-to-end Unsupervised Decision Trees, by Louis Ohl et al.
Kernel KMeans clustering splits for end-to-end unsupervised decision trees
by Louis Ohl, Pierre-Alexandre Mattei, Mickaël Leclercq, Arnaud Droit, Frédéric Precioso
First submitted to arxiv on: 19 Feb 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel end-to-end trained unsupervised binary tree for clustering, called Kauri, is introduced. Unlike previous works that focus on interpreting with trees the result of another clustering algorithm, Kauri performs greedy maximization of the kernel KMeans objective without requiring centroid definition. The model is compared to recent unsupervised trees on multiple datasets and shows identical performance when using a linear kernel, outperforming other models for non-linear kernels. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Kauri is a new way to group similar things together without knowing what they are in advance. It’s like a special kind of tree that helps find patterns in data. The usual approach is to use another clustering method and then explain the results with a tree, but Kauri does it all itself. It works by trying to maximize the kernel KMeans objective in a step-by-step manner without needing any initial definitions. This new model is tested on several datasets and performs just as well as other methods when using a simple linear approach. For more complex patterns, Kauri often outperforms the combination of another clustering method and a decision tree. |
Keywords
* Artificial intelligence * Clustering * Decision tree * Unsupervised
