Generative Local Metric Learning for Kernel Regression

Authors: Yung-Kyun Noh, Masashi Sugiyama, Kee-Eung Kim, Frank Park, Daniel D. Lee

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The proposed algorithm is evaluated using both synthetic and real datasets. For real-data experiments, we used the Delve datasets (Abalone, Bank-8fm, Bank-32fh, CPU), UCI datasets (Community, Naval C, Naval T, Protein, Slice), KEEL datasets (Ailerons, Elevators, Puma32h) [1], and datasets from a previous paper (Pendulum, Pol) [15].
Researcher Affiliation Academia Yung-Kyun Noh Seoul National University, Rep. of Korea nohyung@snu.ac.kr Masashi Sugiyama RIKEN / The University of Tokyo, Japan sugi@k.u-tokyo.ac.kr Kee-Eung Kim KAIST, Rep. of Korea kekim@cs.kaist.ac.kr Frank C. Park Seoul National University, Rep. of Korea fcp@snu.ac.kr Daniel D. Lee University of Pennsylvania, USA ddlee@seas.upenn.edu
Pseudocode Yes Algorithm 1 Generative Local Metric Learning for NW Regression
Open Source Code No The paper does not provide any concrete access to source code, such as a repository link or an explicit statement about code release.
Open Datasets Yes For real-data experiments, we used the Delve datasets (Abalone, Bank-8fm, Bank-32fh, CPU), UCI datasets (Community, Naval C, Naval T, Protein, Slice), KEEL datasets (Ailerons, Elevators, Puma32h) [1], and datasets from a previous paper (Pendulum, Pol) [15].
Dataset Splits No The paper states: 'we choose h from a pre-chosen validation set.' While it mentions using a validation set, it does not provide specific details on the dataset split, such as percentages, sample counts, or the methodology used for splitting.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup Yes Using a Gaussian model with regularized maximum likelihood estimated parameters, we apply a metric which minimizes the bias with a fixed γ = max(|λ1|, |λ2|) 10 2, and we choose h from a pre-chosen validation set.