Semi-Supervised Multi-View Multi-Label Learning with View-Specific Transformer and Enhanced Pseudo-Label
Authors: Quanjiang Li, Tingjin Luo, Mingdie Jiang, Zhangqi Jiang, Chenping Hou, Feijiang Li
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results validate the effectiveness of SMVTEP against state-of-the-art methods. ... Tabel 1 displays the four metrics with LER fixed at 6%, while Fig. 4 illustrates the variation of AP when LER changes. |
| Researcher Affiliation | Academia | 1College of Science, National University of Defense Technology, Changsha 410073, Hunan, China 2Institute of Big Data Science and Industry, Shanxi University, Taiyuan, 030006, Shanxi, China EMAIL, EMAIL, EMAIL, EMAIL, EMAIL, EMAIL. All listed affiliations are universities, and email domains (nudt.edu.cn, sxu.edu.cn) confirm academic institutions. |
| Pseudocode | No | The paper describes the methodology in detail using textual descriptions and figures (like Figure 1 showing the main framework), but it does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks, nor does it present structured steps formatted like code. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code for the described methodology, nor does it provide any links to a code repository. |
| Open Datasets | Yes | Following (Liu et al. 2023b; Zhao et al. 2021), six popular multi-view multi-label datasets are selected in our experiments, i.e., Yeast (Guillaumin 2010), Corel 5k (Duygulu et al. 2002), VOC 2007 (Everingham et al. 2010), ESP Game (Ahn and Dabbish 2004), IAPR TC-12 (Grubinger et al. 2006), and MIR FLICKR (Huiskes and Lew 2008). |
| Dataset Splits | Yes | Each dataset is divided into training, validation and test sets in the ratio of 7:1:2. |
| Hardware Specification | Yes | Our model is implemented by Py Torch on one NVIDIA Ge Force RTX 4090 GPU of 24GB memory. |
| Software Dependencies | No | Our model is implemented by Py Torch on one NVIDIA Ge Force RTX 4090 GPU of 24GB memory. While Py Torch is mentioned as the framework, a specific version number is not provided, and no other software dependencies with version numbers are listed. |
| Experiment Setup | Yes | To simulate the semi-supervised situations, according to the pre-set labeled example ratio (LER), we randomly select LER% instances as labeled ones in the training set. All results are derived from ten independent runs and the final outcomes are presented as average values along with standard deviations. ... In our experiments, we fix the balance parameter α to 9. ... where Margin is used to control the distance between nonmatching sample pairs and we set it to the maximum 1. ... where η1 and η0 are used to control the confidence intervals and fixed at 0.8 in the experiments. ... The heatmap reveals that the performance improves as λ1 gets closer to 100 and λ2 approaches 0.01 |