Eigenvalues Ratio for Kernel Selection of Kernel Methods

Authors: Yong Liu, Shizhong Liao

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Theoretical analysis and experimental results demonstrate that our kernel selection criterion is a good choice for kernel selection.In this section, we will empirically analyze the performance of our proposed eigenvalues ratio criterion (ER).
Researcher Affiliation Academia Yong Liu and Shizhong Liao School of Computer Science and Technology, Tianjin University, Tianjin 300072, P. R. China {yongliu,szliao}@tju.edu.cn
Pseudocode No The paper describes methods and theoretical analysis but does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement or link regarding the availability of open-source code for the described methodology.
Open Datasets Yes The evaluation is made on 8 available public data sets from LIBSVM data seen in Table 1.
Dataset Splits Yes For each data set, we run all methods 30 times with randomly selected 70% of all data for training and the other 30% for testing.The optimal values for the parameters t {1, 4, 16} and η {0.2, 0.6, 1} of ER, and the parameter δ {2i, i = 0, 5, 10, 15, 20} of EP ... are determined by 3-fold cross-validation on the training set.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory, or specific machine configurations) used to run the experiments.
Software Dependencies No The paper mentions using LSSVM but does not provide specific software names with version numbers for its implementation or any other software dependencies.
Experiment Setup Yes We use the popular Gaussian kernels K(x, x ) = exp[-τ||x - x'||^2] as our candidate kernels, τ {2^i, i = -15, -14, . . . , 15}. The learning machine we used is LSSVM. For each kernel selection criterion and each training set, we chose the optimal kernel parameter τ for each fixed regularized parameter λ {0.01, 0.1, 1, 10}. The optimal values for the parameters t {1, 4, 16} and η {0.2, 0.6, 1} of ER, and the parameter δ {2i, i = 0, 5, 10, 15, 20} of EP...