Partition-wise Linear Models

Authors: Hidekazu Oiwa, Ryohei Fujimaki

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate that our partition-wise linear models perform better than or are at least competitive with state-of-the-art region-specific or locally linear models.
Researcher Affiliation Collaboration Hidekazu Oiwa Graduate School of Information Science and Technology The University of Tokyo hidekazu.oiwa@gmail.com Ryohei Fujimaki NEC Laboratories America rfujimaki@nec-labs.com
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks. It refers to details of the algorithm in a 'full version' [14].
Open Source Code No The paper does not provide concrete access to source code for the methodology described in this paper. It only references third-party libraries and code for comparison.
Open Datasets Yes We utilized several standard benchmark datasets from UCI datasets (skin, winequality, census income, twitter, internet ad, energy heat, energy cool, communities), libsvm datasets (a1a, breast cancer), and LIACC datasets (abalone, kinematics, puma8NH, bank8FM).
Dataset Splits Yes Hyperparameters9 were optimized through 10-fold cross validation.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions using 'libsvm package' and 'scikit-learn package' but does not specify their version numbers or other specific software dependencies with versions.
Experiment Setup Yes Hyper-parameters3 were set as λ0 = 0.01 and λP = 0.001. The algorithm was run in 1, 000 iterations.