Version Space Reduction Based on Ensembles of Dissimilar Balanced Perceptrons

Authors: Karen Braga Enes, Saulo Moraes Villela, Raul Fonseca Neto

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct an experimental study on microarray datasets and assess the performance of the proposed method compared to Support Vector Machine and Bayes Point Machine.
Researcher Affiliation Academia 1Department of Computer Science, Universidade Federal de Juiz de Fora, Brazil 2Department of Computer Science, Universidade Federal de Minas Gerais, Brazil
Pseudocode Yes Algorithm 1 describes the Variable Margin Perceptron. Algorithm 2 describes the Version Space Reduction Machine.
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets Yes The datasets can be found in [Glaab et al., 2012], [Zhu et al., 2007] or [Golub et al., 1999].
Dataset Splits Yes for all the experiments, we employ a 10x10-10-fold cross-validation strategy, expect in the SVM case, where we employ a 1x10-10fold.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper mentions software like "Sequential Minimal Optimization" and "PMCMR R package" but does not provide specific version numbers for these dependencies.
Experiment Setup Yes The learning rate parameter was set to = 0.05, for all the Perceptrons. The number of components as 10+1 to avoid ties due to the majority voting strategy. The dissimilarity factor was defined as the highest possible value in such way that the algorithm is able to generate all required components. Table 2 summarizes the dissimilarity factors.