Iterative Exploration-Driven Sparse SDP Clustering via Thompson Sampling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mun, Jongmin, Dubey, Paromita, Fan, Yingying
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912905085059072
author Mun, Jongmin
Dubey, Paromita
Fan, Yingying
author_facet Mun, Jongmin
Dubey, Paromita
Fan, Yingying
contents This paper studies high-dimensional sparse clustering, a combinatorial NP-hard problem arising from the bilinear coupling between cluster assignment and feature selection. We analyze semidefinite programming (SDP) relaxations of $K$-means and establish minimax separation bounds, demonstrating that these relaxations are theoretically robust to feature over-selection: exact recovery is preserved even in the presence of non-informative features. Leveraging this robustness, we propose a block-coordinate ascent framework that alternates between SDP-based clustering and non-conservative feature selection. To address the tendency of deterministic greedy methods to become trapped in local optima, we formulate the feature selection step as a Thompson sampling bandit problem. This approach introduces adaptive memory by aggregating historical variable-selection outcomes into posterior distributions, and selects features via posterior sampling, enabling stochastic exploration that promotes the inclusion of under-explored features and facilitates escape from local maxima. We establish conditions for consistent variable selection and exact clustering recovery, and extend the method to settings with unknown covariance through a scalable, inverse-free estimation procedure. Numerical experiments demonstrate that the proposed memory-driven approach consistently outperforms state-of-the-art sparse clustering methods.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20478
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Iterative Exploration-Driven Sparse SDP Clustering via Thompson Sampling
Mun, Jongmin
Dubey, Paromita
Fan, Yingying
Methodology
90C22, 62H30, 62L05, 68T05, 90C59
G.1.6; I.5.3; I.2.6; G.3
This paper studies high-dimensional sparse clustering, a combinatorial NP-hard problem arising from the bilinear coupling between cluster assignment and feature selection. We analyze semidefinite programming (SDP) relaxations of $K$-means and establish minimax separation bounds, demonstrating that these relaxations are theoretically robust to feature over-selection: exact recovery is preserved even in the presence of non-informative features. Leveraging this robustness, we propose a block-coordinate ascent framework that alternates between SDP-based clustering and non-conservative feature selection. To address the tendency of deterministic greedy methods to become trapped in local optima, we formulate the feature selection step as a Thompson sampling bandit problem. This approach introduces adaptive memory by aggregating historical variable-selection outcomes into posterior distributions, and selects features via posterior sampling, enabling stochastic exploration that promotes the inclusion of under-explored features and facilitates escape from local maxima. We establish conditions for consistent variable selection and exact clustering recovery, and extend the method to settings with unknown covariance through a scalable, inverse-free estimation procedure. Numerical experiments demonstrate that the proposed memory-driven approach consistently outperforms state-of-the-art sparse clustering methods.
title Iterative Exploration-Driven Sparse SDP Clustering via Thompson Sampling
topic Methodology
90C22, 62H30, 62L05, 68T05, 90C59
G.1.6; I.5.3; I.2.6; G.3
url https://arxiv.org/abs/2505.20478