Uniform concentration bounds toward a unified framework for robust clustering
D Paul, S Chakraborty, S Das… - Advances in Neural …, 2021 - proceedings.neurips.cc
Recent advances in center-based clustering continue to improve upon the drawbacks of
Lloyd's celebrated $ k $-means algorithm over $60 $ years after its introduction. Various …
Lloyd's celebrated $ k $-means algorithm over $60 $ years after its introduction. Various …
Bregman power k-means for clustering exponential family data
A Vellal, S Chakraborty, JQ Xu - International Conference on …, 2022 - proceedings.mlr.press
Recent progress in center-based clustering algorithms combats poor local minima by implicit
annealing through a family of generalized means. These methods are variations of Lloyd's …
annealing through a family of generalized means. These methods are variations of Lloyd's …
On consistent entropy-regularized k-means clustering with feature weight learning: Algorithm and statistical analyses
S Chakraborty, D Paul, S Das - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Clusters in real data are often restricted to low-dimensional subspaces rather than the entire
feature space. Recent approaches to circumvent this difficulty are often computationally …
feature space. Recent approaches to circumvent this difficulty are often computationally …
Robust and Automatic Data Clustering: Dirichlet Process meets Median-of-Means
S Basu, JR Choudhury, D Paul, S Das - arXiv preprint arXiv:2311.15384, 2023 - arxiv.org
Clustering stands as one of the most prominent challenges within the realm of unsupervised
machine learning. Among the array of centroid-based clustering algorithms, the classic $ k …
machine learning. Among the array of centroid-based clustering algorithms, the classic $ k …
On uniform concentration bounds for bi-clustering by using the Vapnik–Chervonenkis theory
S Chakraborty, S Das - Statistics & Probability Letters, 2021 - Elsevier
Bi-clustering refers to the task of partitioning the rows and columns of a data matrix
simultaneously. Although empirically useful, the theoretical aspects of bi-clustering …
simultaneously. Although empirically useful, the theoretical aspects of bi-clustering …
Upper bound estimations of misclassification rate in the heteroscedastic clustering model with sub‐Gaussian noises
Z Zhang, J Wang - Stat, 2023 - Wiley Online Library
Clustering is an important tool in statistics, machine learning and applied mathematics. This
paper considers the clustering model Y= μ l T+ Z∈ ℝ p× n, where the noise matrix Z consists …
paper considers the clustering model Y= μ l T+ Z∈ ℝ p× n, where the noise matrix Z consists …
Robust Linear Predictions: Analyses of Uniform Concentration, Fast Rates and Model Misspecification
S Chakraborty, D Paul, S Das - arXiv preprint arXiv:2201.01973, 2022 - arxiv.org
The problem of linear predictions has been extensively studied for the past century under
pretty generalized frameworks. Recent advances in the robust statistics literature allow us to …
pretty generalized frameworks. Recent advances in the robust statistics literature allow us to …
t-Divergence: A New Divergence Measure with Application to Robust Statistics & Clustering
D Paul, S Chakraborty, S Das - The Second Tiny Papers Track at ICLR … - openreview.net
This paper introduces the $ t $-divergence, a novel divergence measure associated with the
inverse tangent function. We investigate its intriguing consistent and outlier-robust features …
inverse tangent function. We investigate its intriguing consistent and outlier-robust features …