Near-Optimal Algorithms for Explainable k-Medians and k-Means

Konstantin Makarychev, Liren Shan
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:7358-7367, 2021.

Abstract

We consider the problem of explainable k-medians and k-means introduced by Dasgupta, Frost, Moshkovitz, and Rashtchian (ICML 2020). In this problem, our goal is to find a \emph{threshold decision tree} that partitions data into k clusters and minimizes the k-medians or k-means objective. The obtained clustering is easy to interpret because every decision node of a threshold tree splits data based on a single feature into two groups. We propose a new algorithm for this problem which is ˜O(logk) competitive with k-medians with ℓ1 norm and ˜O(k) competitive with k-means. This is an improvement over the previous guarantees of O(k) and O(k2) by Dasgupta et al (2020). We also provide a new algorithm which is O(log\nicefrac32k) competitive for k-medians with ℓ2 norm. Our first algorithm is near-optimal: Dasgupta et al (2020) showed a lower bound of Ω(logk) for k-medians; in this work, we prove a lower bound of ˜Ω(k) for k-means. We also provide a lower bound of Ω(logk) for k-medians with ℓ2 norm.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-makarychev21a, title = {Near-Optimal Algorithms for Explainable k-Medians and k-Means}, author = {Makarychev, Konstantin and Shan, Liren}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {7358--7367}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/makarychev21a/makarychev21a.pdf}, url = {https://proceedings.mlr.press/v139/makarychev21a.html}, abstract = {We consider the problem of explainable $k$-medians and $k$-means introduced by Dasgupta, Frost, Moshkovitz, and Rashtchian (ICML 2020). In this problem, our goal is to find a \emph{threshold decision tree} that partitions data into $k$ clusters and minimizes the $k$-medians or $k$-means objective. The obtained clustering is easy to interpret because every decision node of a threshold tree splits data based on a single feature into two groups. We propose a new algorithm for this problem which is $\tilde O(\log k)$ competitive with $k$-medians with $\ell_1$ norm and $\tilde O(k)$ competitive with $k$-means. This is an improvement over the previous guarantees of $O(k)$ and $O(k^2)$ by Dasgupta et al (2020). We also provide a new algorithm which is $O(\log^{\nicefrac{3}{2}} k)$ competitive for $k$-medians with $\ell_2$ norm. Our first algorithm is near-optimal: Dasgupta et al (2020) showed a lower bound of $\Omega(\log k)$ for $k$-medians; in this work, we prove a lower bound of $\tilde\Omega(k)$ for $k$-means. We also provide a lower bound of $\Omega(\log k)$ for $k$-medians with $\ell_2$ norm.} }
Endnote
%0 Conference Paper %T Near-Optimal Algorithms for Explainable k-Medians and k-Means %A Konstantin Makarychev %A Liren Shan %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-makarychev21a %I PMLR %P 7358--7367 %U https://proceedings.mlr.press/v139/makarychev21a.html %V 139 %X We consider the problem of explainable $k$-medians and $k$-means introduced by Dasgupta, Frost, Moshkovitz, and Rashtchian (ICML 2020). In this problem, our goal is to find a \emph{threshold decision tree} that partitions data into $k$ clusters and minimizes the $k$-medians or $k$-means objective. The obtained clustering is easy to interpret because every decision node of a threshold tree splits data based on a single feature into two groups. We propose a new algorithm for this problem which is $\tilde O(\log k)$ competitive with $k$-medians with $\ell_1$ norm and $\tilde O(k)$ competitive with $k$-means. This is an improvement over the previous guarantees of $O(k)$ and $O(k^2)$ by Dasgupta et al (2020). We also provide a new algorithm which is $O(\log^{\nicefrac{3}{2}} k)$ competitive for $k$-medians with $\ell_2$ norm. Our first algorithm is near-optimal: Dasgupta et al (2020) showed a lower bound of $\Omega(\log k)$ for $k$-medians; in this work, we prove a lower bound of $\tilde\Omega(k)$ for $k$-means. We also provide a lower bound of $\Omega(\log k)$ for $k$-medians with $\ell_2$ norm.
APA
Makarychev, K. & Shan, L.. (2021). Near-Optimal Algorithms for Explainable k-Medians and k-Means. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:7358-7367 Available from https://proceedings.mlr.press/v139/makarychev21a.html.

Related Material