Regression with reject option and application to kNN

Authors: Ahmed Zaoui, Christophe Denis, Mohamed Hebiri

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, a numerical study is performed to illustrate the benefit of using the proposed procedure.
Researcher Affiliation Academia Ahmed Zaoui LAMA, Université Gustave Eiffel ahmed.zaoui@univ-eiffel.fr Christophe Denis LAMA, Université Gustave Eiffel MIA-Paris, Agro Paris Tech, INRAE, Université Paris-Saclay christophe.denis@univ-eiffel.fr Mohamed Hebiri LAMA, Université Gustave Eiffel CREST, ENSAE, Institut Polytechnique de Paris mohamed.hebiri@univ-eiffel.fr
Pseudocode No The paper does not contain any clearly labeled 'Pseudocode' or 'Algorithm' sections, nor any structured code-like blocks.
Open Source Code Yes The code used for the implementation of the plug-in ε-predictor can be found at https://github.com/Zaoui Amed/Neurips2020_ Reject Option.
Open Datasets Yes The performance is evaluated on two benchmark datasets: QSAR aquatic toxicity and Airfoil Self Noise coming from the UCI database.
Dataset Splits Yes For all datasets, we split the data into three parts (50 % train labeled, 20 % train unlabeled, 30 % test).
Hardware Specification No The paper does not provide specific hardware details (such as GPU or CPU models, or memory specifications) used for running the experiments.
Software Dependencies No For random forests and svm procedures, we used respectively the R packages random Forest and e1071 with default parameters.
Experiment Setup Yes We employ the 10-fold cross-validation to select the parameter k {5, 10, 15, 20, 30, 50, 70, 100, 150} of the k NN algorithm.