A General Formulation for Safely Exploiting Weakly Supervised Data

Authors: Lan-Zhe Guo, Yu-Feng Li

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on multiple weakly supervised learning tasks such as label noise learning, domain adaptation and semi-supervised learning validate the effectiveness of our proposed algorithms.
Researcher Affiliation Academia Lan-Zhe Guo, Yu-Feng Li National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China {guolz, liyf}@lamda.nju.edu.cn
Pseudocode No No structured pseudocode or algorithm blocks were found in the paper.
Open Source Code Yes We call our proposal as SAFEW (SAFE Weakly supervised learning)1. 1http://lamda.nju.edu.cn/code/SAFEW.ashx
Open Datasets Yes We conduct experimental comparison for label noise learning tasks on a number of frequently-used classification datasets2, i.e., Australian, Breast-Cancer, Diabetes, Digit1, Heart, Ionosphere, USPS and Splice. 2https://www.csie.ntu.edu.tw/ cjlin/libsvmtools/datasets/
Dataset Splits Yes For each data set, 80% of instances are used for training and the rest ones are used for testing.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, processor types, memory amounts) used for running experiments were mentioned in the paper.
Software Dependencies No The paper mentions 'glmfit function in Matlab' and 'Libsvm package (Chang and Lin 2011)' but does not provide specific version numbers for these software components or any other dependencies.
Experiment Setup Yes For k-NN method, k is set to 3.