Sparse Shrunk Additive Models

Authors: Guodong Liu, Hong Chen, Heng Huang

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on both synthetic and real-world data demonstrate the effectiveness of the proposed approach.
Researcher Affiliation Collaboration 1Department of Electrical and Computer Engineering, University of Pittsburgh, PA, USA 2Department of Mathematics and Statistics, College of Science, Huazhong Agricultural University, Wuhan, China 3JD Finance America Corporation, Mountain View, CA, USA.
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access to source code for the methodology described.
Open Datasets Yes The data sets are from UCI repository (http://archive.ics.uci.edu/ml) and (Tu et al., 2012), which include Insulin (n = 50, m = 256), Skillcraft (n = 18, m = 1700), Airfoil (n = 40, m = 750), Forestfire (n = 10, m = 211), Housing (n = 12, m = 256), CCPP (n = 59, m = 2000), Music (n = 90, m = 1000), Telemonit (n = 19, m = 1000).
Dataset Splits Yes The regularization parameter λ is chosen via five-fold cross validation with respect to the mean square error (MSE).
Hardware Specification No The paper does not provide any specific hardware details used for running its experiments.
Software Dependencies No The paper mentions implementing the method via 'accelerated proximal gradient methods (Nesterov, 2013)' but does not list any specific software libraries or their version numbers.
Experiment Setup Yes The regularization parameter λ is chosen via five-fold cross validation with respect to the mean square error (MSE). ... each kernel on X (j) is generated from Gaussian kernel. For example, when x(j) s = (xs1, xs2) and x(j) t = (xt1, xt2), the shrunk kernel K(j)(x(j) s , x(j) t ) = exp{ (xs1 xt1)2 / (2µ2 1) }exp{ (xs2 xt2)2 / (2µ2 2) }, where µi = 4.5σim −1/10 and σi is the standard deviation on i-th coordination. ... The final output is y = f (x) + ϵ, where ϵ N(0, 0.25).