Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

The Product Cut

Authors: Thomas Laurent, James von Brecht, Xavier Bresson, arthur szlam

NeurIPS 2016 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conclude with an experimental evaluation and comparison of the algorithm on real world data sets to validate these claims.
Researcher Affiliation Collaboration Xavier Bresson Nanyang Technological University Singapore EMAIL Thomas Laurent Loyola Marymount University Los Angeles EMAIL Arthur Szlam Facebook AI Research New York EMAIL James H. von Brecht California State University, Long Beach Long Beach EMAIL
Pseudocode Yes Algorithm 1 Randomized SLP for PCut
Open Source Code Yes 1The code is available at https://github.com/xbresson/pcut
Open Datasets Yes We provide experimental results on four text data sets (20NEWS, RCV1, WEBKB4, CITESEER) and four data sets containing images of handwritten digits (MNIST, PENDIGITS, USPS, OPTDIGITS).
Dataset Splits No The paper uses datasets for evaluation but does not specify training, validation, or test splits.
Hardware Specification No The paper states that experiments were performed on the 'same architecture' but does not provide specific hardware details (e.g., CPU/GPU models, memory).
Software Dependencies No The paper refers to various algorithms and methods (e.g., NCut, NMFR, Algebraic Multigrid) but does not provide specific software names with version numbers for reproducibility.
Experiment Setup Yes For the PCut algorithm, we use α = .9 when defining Ωα. Also, in order to illustrate the tradeoff when selecting the rate at which the number of enforced constraints sk increases, we report accuracy results for the linear rates sk = 10 4 n := λ1 and sk = 5 10 4 n := λ2 where n denotes the total number of vertices in the data set.