Generating CP-Nets Uniformly at Random

Authors: Thomas Allen, Judy Goldsmith, Hayden Justice, Nicholas Mattei, Kayla Raines

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

Reproducibility Variable Result LLM Response
Research Type Experimental We provide a novel algorithm for provably generating acyclic CP-nets uniformly at random. Our method is computationally efficient and allows for multi-valued domains and arbitrary bounds on the indegree in the dependency graph. [...] We have implemented our method in C++ using the Gnu MP library (Granlund et al. 2014), allowing generation of thousands of CP-nets per second.
Researcher Affiliation Academia Thomas E. Allen University of Kentucky Lexington, Kentucky, USA teal223@g.uky.edu Judy Goldsmith University of Kentucky Lexington, Kentucky, USA goldsmit@cs.uky.edu Hayden E. Justice The Gatton Academy, WKU Bowling Green, Kentucky, USA hayden.justice259@topper.wku.edu Nicholas Mattei Data61 and UNSW Sydney, Australia nicholas.mattei@nicta.com.au Kayla Raines University of Kentucky Lexington, Kentucky, USA kayla.raines@live.com
Pseudocode Yes Algorithm 1: DAGCODE-TO-DAG [...] Algorithm 2: ALL-DAGS [...] Algorithm 3: BUILD-CP-NET [...] Algorithm 4: ALL-CP-NETS [...] Algorithm 5: COMPUTE-DISTRIBUTION [...] Algorithm 6: RANDOM-CP-NET
Open Source Code Yes Our code is available at http://cs.uky.edu/ goldsmit/papers/ Generating CPnet Code.html.
Open Datasets No The paper discusses generating CP-nets and their properties, but does not describe experiments involving training datasets for models or algorithms.
Dataset Splits No The paper does not explicitly provide validation dataset splits. Its focus is on a generation algorithm, not empirical model training.
Hardware Specification No The paper states, 'We have implemented our method in C++ [...] allowing generation of thousands of CP-nets per second,' but does not specify the hardware used for this implementation or performance measurement.
Software Dependencies Yes We have implemented our method in C++ using the Gnu MP library (Granlund et al. 2014).
Experiment Setup No The paper describes its algorithm for generating CP-nets but does not detail an experimental setup with specific hyperparameters or system-level training settings, as it is not training a machine learning model.