Fast Lasso Algorithm via Selective Coordinate Descent

Authors: Yasuhiro Fujiwara, Yasutoshi Ida, Hiroaki Shiokawa, Sotetsu Iwamura

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We performed experiments on the datasets of DNA, Protein, Reuters, TDT2, and Newsgroups to show the efficiency and effectiveness of our approach.
Researcher Affiliation Collaboration NTT Software Innovation Center, 3-9-11 Midori-cho Musashino-shi, Tokyo, 180-8585, Japan Center for Computational Science, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki, 305-8573, Japan
Pseudocode Yes Algorithm 1 Sling
Open Source Code No The paper does not provide an explicit statement or link for open-source code specific to the Sling methodology described in the paper.
Open Datasets Yes We performed experiments on the datasets of DNA, Protein, Reuters, TDT2, and Newsgroups to show the efficiency and effectiveness of our approach. Details of the datasets are shown in Chih-Jen Lin s webpage2 and Deng Cai s webpage3.
Dataset Splits Yes This experiment performed leave-one-out cross validation in evaluating the prediction error in terms of the squared loss for the response.
Hardware Specification Yes We conducted all experiments on a Linux 2.70 GHz Intel Xeon server.
Software Dependencies No The paper states 'We implemented all approaches using GCC' but does not provide specific version numbers for GCC or any other software libraries.
Experiment Setup Yes We set λ1 = 1/n maxi | xi, y | and λK = 0.001λ1 by following the previous paper (Friedman, Hastie, and Tibshirani 2010). We constructed a sequence of K scores of tuning parameters decreasing from λ1 to λK on a log scale where K = 50.