Solving Interpretable Kernel Dimensionality Reduction
Authors: Chieh Wu, Jared Miller, Yale Chang, Mario Sznaier, Jennifer Dy
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 4 Experiments. The experiment includes 5 real datasets of commonly encountered data types. Wine [33] consists of continuous data while the Cancer dataset [34] features are discrete. The Face dataset [35] is a standard dataset used for alternative clustering; it includes images of 20 people in various poses. The MNIST [36] dataset includes images of handwritten characters. |
| Researcher Affiliation | Academia | Electrical and Computer Engineering Dept., Northeastern University, Boston, MA |
| Pseudocode | Yes | Algorithm 1 ISM Algorithm Input : Data X, kernel, Subspace Dimension q Output : Projected subspace W |
| Open Source Code | Yes | To support reproducible results, the source code is made publicly available on https://github.com/chieh-neu/ISM_supervised_DR. |
| Open Datasets | Yes | Wine [33] consists of continuous data while the Cancer dataset [34] features are discrete. The Face dataset [35] is a standard dataset used for alternative clustering; it includes images of 20 people in various poses. The MNIST [36] dataset includes images of handwritten characters. |
| Dataset Splits | Yes | For supervised dimension reduction, we perform SVM on XW using 10-fold cross validation. |
| Hardware Specification | Yes | All experiments were conducted on Dual Intel Xeon E5-2680 v2 @ 2.80GHz, with 20 total cores. |
| Software Dependencies | No | All sources are written in Python using Numpy and Sklearn [41; 42]. Specific version numbers for Python, Numpy, or Sklearn are not provided. |
| Experiment Setup | Yes | The median of the pair-wise Euclidean distance is used as σ for all experiments using the Gaussian kernel. Degree of 3 is used for all polynomial kernels. The dimension of subspace q is set to the number of classes/clusters. The convergence threshold δ is set to 0.01. |