Likelihood Adjusted Semidefinite Programs for Clustering Heterogeneous Data

Authors: Yubo Zhuang, Xiaohui Chen, Yun Yang

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our numeric experiments demonstrate that i LASDP can achieve lower mis-clustering errors over several widely used clustering methods including K-means, SDP and EM algorithms. In this section, we compare the performance of several widely used clustering methods on two real datasets.
Researcher Affiliation Academia 1Department of Statistics, University of Illinois at Urbana Champaign. 2Department of Mathematics, University of Southern California. Correspondence to: Yubo Zhuang <yubo2@illinois.edu>, Xiaohui Chen <xiaohuic@usc.edu>, Yun Yang <yy84@illinois.edu>.
Pseudocode Yes Algorithm 1 The iterative likelihood adjusted SDP (i LASDP) algorithm
Open Source Code No The paper does not contain any explicit statement or link providing access to the source code for the methodology described.
Open Datasets Yes The first one is handwriting digits dataset MNIST, the second one is CIFAR-10 image dataset and the last one is Landsat dataset from the UCI machine learning repository.
Dataset Splits No The paper mentions the use of a "test set" for various datasets (MNIST, CIFAR-10, Landsat) but does not provide explicit details about training or validation splits (e.g., percentages or sample counts for all splits), only the size of the test sets.
Hardware Specification No The paper does not specify any particular hardware (e.g., GPU models, CPU models, memory, or cloud instance types) used for running the experiments.
Software Dependencies No The paper mentions "hierarchical clustering from mclust package in R" but does not provide a specific version number for this or any other software dependency.
Experiment Setup Yes For all the dimension reduction procedures used in the simulation experiments, we perform step 1-7 in Algorithm 2 followed by Algorithm 3 with input parameters α = 0.7, C = 1010, p0 = 2K, p1 = 15 ϵ = 10 2, S = 50.