Overview

Unsupervised clustering studies how to divide unlabeled data into meaningful groups, supporting tasks such as scientific discovery, data exploration, and anomaly analysis. Methods range from k-means and hierarchical clustering to density-based and spectral approaches, with newer work learning representations, feature weights, or shared structure across views and time. Key challenges include choosing the number of clusters, handling noise, missing data, and unequal cluster shapes, and making results stable, interpretable, and scalable. Deep methods and language-model-assisted approaches can help in some settings, but benchmarks show that conventional algorithms often remain competitive, so method choice and reliable evaluation depend strongly on the data and task.

Latest papers

122 papers6 papers in the last four weeks, down 45% on the four weeks before. 0.0% of all new papers.

Jan 16, 2024

Semidefinite programming relaxations and debiasing for MAXCUT-based clustering

In this paper, we consider the problem of partitioning a small data sample of size nn drawn from a mixture of 22 sub-gaussian distributions in Rp\mathbb{R}^p. We consider semidefinite programming relaxations of an integer quadratic program that is formulated essentially as finding the maximum cut on a graph, where edge weights in the cut represent dissimilarity scores between two nodes based on their pp features. We define the signal-to-noise ratio (SNR) as s2:=min⁡{npγ2,Δ2}s^2 := \min\{n p γ^2, Δ^2\}, where Δ2:=pγΔ^2 := p γ denotes the ℓ22\ell_2^2 distance between the two cluster centers. Our contributions are twofold. First, we provide a unified framework for analyzing three computationally efficient algorithms: SDP1, BalancedSDP, and Spectral clustering, yielding universal polynomial-rate misclassification guarantees for all three algorithms. Moreover, our theory allows for partial recovery (success rate <100%< 100\%) as long as s2s^2 is lower bounded by a constant. Second, we prove that the misclassification errors for SDP1 and BalancedSDP decay exponentially with respect to the SNR s2s^2 and the BalancedSDP requires no explicit debiasing when the two clusters have equal sizes. To our knowledge, this is the first time such results are obtained for semidefinite relaxations of MAX CUT in population clustering. We provide simulation evidence illuminating the theoretical predictions.
Jun 22, 2023

Reinforcement Federated Learning Method Based on Adaptive OPTICS Clustering

Federated learning is a distributed machine learning technology, which realizes the balance between data privacy protection and data sharing computing. To protect data privacy, feder-ated learning learns shared models by locally executing distributed training on participating devices and aggregating local models into global models. There is a problem in federated learning, that is, the negative impact caused by the non-independent and identical distribu-tion of data across different user terminals. In order to alleviate this problem, this paper pro-poses a strengthened federation aggregation method based on adaptive OPTICS clustering. Specifically, this method perceives the clustering environment as a Markov decision process, and models the adjustment process of parameter search direction, so as to find the best clus-tering parameters to achieve the best federated aggregation method. The core contribution of this paper is to propose an adaptive OPTICS clustering algorithm for federated learning. The algorithm combines OPTICS clustering and adaptive learning technology, and can effective-ly deal with the problem of non-independent and identically distributed data across different user terminals. By perceiving the clustering environment as a Markov decision process, the goal is to find the best parameters of the OPTICS cluster without artificial assistance, so as to obtain the best federated aggregation method and achieve better performance. The reliability and practicability of this method have been verified on the experimental data, and its effec-tiveness and superiority have been proved.