Fast Low-rank Metric Learning for Large-scale and High-dimensional Data

Authors: Han Liu, Zhizhong Han, Yu-Shen Liu, Ming Gu

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The outperforming experimental results show that our method is with high accuracy and much faster than the state-of-the-art methods under several benchmarks with large numbers of high-dimensional data.
Researcher Affiliation Academia Han Liu , Zhizhong Han , Yu-Shen Liu , Ming Gu School of Software, Tsinghua University, Beijing, China BNRist & KLISS, Beijing, China Department of Computer Science, University of Maryland, College Park, USA
Pseudocode Yes The outline of the FLRML algorithm is shown in Algorithm 1. It can be mainly divided into four stages: SVD preprocessing (line 2), constant initializing (line 3), variable initializing (lines 4 and 5), and the iterative optimization (lines 6 to 11).
Open Source Code Yes Code has been made available at https://github.com/highan911/FLRML.
Open Datasets Yes The methods are evaluated on eight datasets with high dimensions or large numbers of samples: three datasets NG20, RCV1-4 and TDT2-30 derived from three text collections respectively [35, 36]; one handwritten characters dataset MNIST [37]; four voxel datasets of 3D models M10-16, M10-100, M40-16, and M40-100 with different resolutions in 163 and 1003 dimensions, respectively, generated from Model Net10 and Model Net40 [38] which are widely used in 3D shape understanding [39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50].
Dataset Splits No The dimensions D, the number of training samples n, the number of test samples ntest, and the number of categories ncat of all the datasets are listed in Table 1.
Hardware Specification Yes All the experiments are performed on the Matlab R2015a platform on a PC with 3.60GHz processor and 16GB of physical memory.
Software Dependencies Yes All the experiments are performed on the Matlab R2015a platform on a PC with 3.60GHz processor and 16GB of physical memory.
Experiment Setup Yes For all these methods, the rank for SVD is set as r = min(rank(X), 3000). For the other methods, 5 triplets are randomly generated for each sample, which is also used as 5 positive pairs and 5 negative pairs for the methods using pairwise constraints. The accuracy is evaluated by a 5-NN classifier using the output metric of each method. For each low-rank metric learning method, the rank constraint for M is set as d = 100. So we use m/ ly = 1 in the experiments in Table 3 and Table 4. So in Table 4, all the results are obtained with Nt = 80 and T = 20.