Learning in Reproducing Kernel Kreı̆n Spaces

Authors: Dino Oglic, Thomas Gaertner

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

Reproducibility Variable Result LLM Response
Research Type Experimental The approach is evaluated empirically using indefinite kernels defined on structured as well as vectorial data. The empirical results demonstrate a superior performance of our approach over the state-of-the-art baselines.
Researcher Affiliation Academia 1School of Computer Science, University of Nottingham, UK 2Institut f ur Informatik III, Universit at Bonn, Germany.
Pseudocode No The paper describes algorithms and derivations mathematically but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statements about releasing its source code or links to a code repository.
Open Datasets Yes All datasets have been downloaded from the LIBSVM library (Chang & Lin, 2011). ... using a set of benchmark datasets for learning with indefinite kernels (Duin & Pekalska, 2009).
Dataset Splits Yes We measure the effectiveness of a baseline/method using the average root mean squared error, computed after performing 10 fold outer cross-validation. ... 10 fold stratified cross-validation.
Hardware Specification No The paper mentions 'University of Nottingham High Performance Computing Facility' in the acknowledgements, but it does not specify any particular hardware details such as CPU/GPU models or memory.
Software Dependencies No The paper mentions 'L-BFGS-B minimization procedure' and 'LIBSVM library (Chang & Lin, 2011)' but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes In Section 3.3, we derive the gradient of an optimal solution to the risk minimization problem with respect to the hyperparameters of the model (e.g., the regularization parameters, hypersphere radius, and/or kernel-specific parameters). ... A detailed description of the experimental setup can be found in Appendix C.