The Weighted Kendall and High-order Kernels for Permutations
Authors: Yunlong Jiao, Jean-Philippe Vert
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we demonstrate the use of the proposed weighted kernels compared with the standard Kendall kernel for classification on a real dataset from the European Union survey Eurobarometer 55.2 (Christensen, 2010). |
| Researcher Affiliation | Academia | 1University of Oxford, Oxford, UK 2MINES Paris Tech & Institut Curie & Ecole Normale Sup erieure, PSL Research University, Paris, France. |
| Pseudocode | Yes | The proof is constructive and deferred to the supplements, where we also detail the pseudo-code for a fast computational algorithm. |
| Open Source Code | Yes | Code to reproduce all the experiments in the present paper is available at https://github.com/Yunlong Jiao/ weightedkendall. |
| Open Datasets | Yes | real dataset from the European Union survey Eurobarometer 55.2 (Christensen, 2010). |
| Dataset Splits | No | we randomly sub-sampled a training set of 400 participants and a test set of 100 participants. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for the experiments. |
| Software Dependencies | No | The paper mentions 'C++/R implementation' and 'kernel SVMs' but does not provide specific version numbers for software components or libraries used. |
| Experiment Setup | No | The paper states 'we chose to fit kernel SVMs with different kernels' but does not specify concrete experimental setup details such as hyperparameter values (e.g., learning rate, batch size, regularization) for these models. |