Neuc-MDS: Non-Euclidean Multidimensional Scaling Through Bilinear Forms

Authors: Chengyuan Deng, Jie Gao, Kevin Lu, Feng Luo, Hongbin Sun, Cheng Xin

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental This section presents experimental results of Neuc-MDS and Neuc-MDS+. First, we evaluate the performance on dissimilarity error of two proposed algorithms comparing with closelyrelated baselines on three metrics: STRESS, distortion and additive error. Empirically we evaluate Neuc-MDS and Neuc-MDS+ on ten diverse datasets encompassing different domains. The experiment results show that both methods substantially outperform previous baselines on STRESS and average distortion, and fully resolve the issue of dimensionality paradox in c MDS.
Researcher Affiliation Academia Rutgers University. {cd751,jg1555,kll160,fluo,hongbin.sun,cx122}@rutgers.edu
Pseudocode Yes Algorithm 1: Non-Euclidean Multidimensional Scaling; Algorithm 2: Eigenvalues Selection for Neuc-MDS; Algorithm 3: Eigenvalue Selection for Generalized Neuc-MDS
Open Source Code Yes Our codes are available on Github2. 2https://github.com/KLu9812/MDSPlus
Open Datasets Yes We also test three celebrated image datasets: MNIST, Fashion-MNIST and CIFAR-10. The dissimilarity matrix for each dataset captures 1000 images randomly sampled for each class. We use three measures in [38] to calculate the dissimilarities. For genomics data, we include 5 datasets from the Curated Microarray Database (Cu Mi Da) [17]
Dataset Splits No The paper evaluates performance on various datasets but does not explicitly detail train/validation/test splits, specific splitting methodologies, or cross-validation setups.
Hardware Specification Yes Our experiments are implemented with Intel Core i9 CPU of 32GB memory, no GPU is required and the execution time is no longer than 30 seconds.
Software Dependencies No The paper mentions using 'Scikit-learn: Machine learning in Python.' [29], but does not provide specific version numbers for this or any other software dependencies.
Experiment Setup Yes With k as the target dimension, for synthetic datasets and images, we set k = 100, for genomics data, k = 20. The details are presented in Table 2. ... We use three measures in [38] to calculate the dissimilarities. ... For genomics datasets, we use the entropic affinities commonly used in t-SNE methods.