Label-Sensitive Task Grouping by Bayesian Nonparametric Approach for Multi-Task Multi-Label Learning

Authors: Xiao Zhang, Wenzhong Li, Vu Nguyen, Fuzhen Zhuang, Hui Xiong, Sanglu Lu

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate the model performance on three public multi-task multi-label data sets, and the results show that LABTAG outperforms the compared baselines with a significant margin. 4 Experimental Evaluation
Researcher Affiliation Academia 1 State Key Laboratory for Novel Software Technology, Nanjing University, China 2 Center for Pattern Recognition and Data Analytics, Deakin University, Australia 3 Key Lab of IIP of CAS, Institute of Computing Technology, CAS Beijing, China 4 Management Science & Information Systems, Rutgers University, USA
Pseudocode Yes Algorithm 1 The generative process of the proposed LABTAG model
Open Source Code No The paper mentions utilizing and downloading code for baseline models ('MEKA toolbox', 'Matlab code from the author s website for BNMC and ML-KNN'), but does not provide concrete access to the source code for the proposed LABTAG model.
Open Datasets Yes We use three public data sets for the performance evaluation, including Trip Advisor, LDOS-Co Mo Da and Enron Corpus data sets... Trip Advisor and LDOS-Co Mo Da are two context-aware data sets, which are used for context recommendation systems [Zheng et al., 2014]... Finally, the Enron Email Corpus contains email information (email content and recipients) from Enron [Klimt and Yang, 2004; Carvalho and Cohen, 2007].
Dataset Splits Yes In each task, 50% data are used for training and the remaining 50% for test.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions software like 'MEKA toolbox' and 'Matlab code' used for baselines, but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes The hyper-parameter settings of LABTAG model are as follows: α = 1, δ = 0.01, ϖ = 0.07, β = 1, µ0 = 0, Σ0 = 10I. The truncation threshold is set as 0.001 #Train and the learning rate is set as 0.01.