Distributed Semi-supervised Learning with Kernel Ridge Regression
Authors: Xiangyu Chang, Shao-Bo Lin, Ding-Xuan Zhou
JMLR 2017 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we report experimental studies to justify the statements in Section 2. We employ two criteria for the comparison. The first criterion is the global error (GE) which is the mean square error of a testing set with N = |D| examples used in the training flow. GE provides a baseline to assess the performance of DSKRR. The second criterion is the average error (AE) which is the mean square error of algorithm (4). Regularization parameters in all experiments are selected by the 5-fold cross-validation. |
| Researcher Affiliation | Academia | Xiangyu Chang EMAIL Center of Data Science and Information Quality School of Management Xi an Jiaotong University, Xi an, China Shao-Bo Lin EMAIL Department of Statistics Wenzhou University, Wenzhou, China Ding-Xuan Zhou EMAIL Department of Mathematics City University of Hong Kong Tat Chee Avenue, Kowloon, Hong Kong, China |
| Pseudocode | No | The paper defines algorithms (1) and (4) using mathematical notation, not structured pseudocode blocks or clearly labeled algorithm sections. For instance, algorithm (4) is defined as: |D j| |D |f D j ,λ. No explicit 'Algorithm' or 'Pseudocode' sections are present. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code for the described methodology, nor does it provide links to any code repositories. There are no mentions of supplementary materials containing code. |
| Open Datasets | Yes | In this part, we focus on the Million Song data (Bertin-Mahieux et al., 2011) that describes a learning task of predicting the year in which a song is released based on audio features associated with the song. |
| Dataset Splits | Yes | The dataset consists of 463,715 training examples and 51,630 test examples. ... Regularization parameters in all experiments are selected by the 5-fold cross-validation. |
| Hardware Specification | No | The paper does not provide specific hardware details such as CPU models, GPU types, or memory specifications used for running the experiments. It only describes the computational tasks without mentioning the underlying hardware. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers, such as programming languages, libraries, or frameworks used for implementation. It only mentions general techniques like 'kernel ridge regression'. |
| Experiment Setup | Yes | Regularization parameters in all experiments are selected by the 5-fold cross-validation. ... Finally, we use the Gaussian kernel K(x, x ) = exp n x x 2 2 2β2 o in our experiments with bandwidth parameter β = 6 and regularization parameter λ = N 1/2. |