Approximate and Exact Enumeration of Rule Models

Authors: Satoshi Hara, Masakazu Ishihata

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We then confirm our theoretical results through experiments on real-world data. We also show that, by using the proposed enumeration algorithms, we can find several different models of almost equal quality. and In this section, we conduct experiments to evaluate the proposed algorithms.
Researcher Affiliation Academia 1) Osaka University, Osaka, Japan 2) JST, ERATO, Kawarabayashi Large Graph Project 3) Hokkaido University, Hokkaido, Japan
Pseudocode Yes Algorithm 1 Approximate Enumeration Algorithm and Algorithm 2 Exact Enumeration Algorithm
Open Source Code No The paper does not contain an explicit statement or link to the open-source code for the methodology described in this paper. It only references third-party repositories for tools used.
Open Datasets Yes To evaluate the algorithms, we used two classification datasets of categorical data: COMPAS (Larson et al. 2016) and Mushroom (Lichman 2013).
Dataset Splits No For COMPAS, it mentions '6,489 training samples and 721 test samples'. For Mushroom, it states 'randomly split the samples into 6,499 (80%) training samples and 1,625 (20%) test samples'. While training and test splits are mentioned, there is no explicit mention of a validation split.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper states 'Algorithm 1 was implemented in Python 3.5, while Algorithm 2 was implemented in C.' It does not provide specific version numbers for any libraries or frameworks used within Python or C.
Experiment Setup Yes For CORELS, we used the configurations recommended in the github repository with regularization parameter ρ = 0.015. We set the length of the rule sets to be I = 2. and In the experiment, we set the length of the rule sets to be I = 4.