One Positive Label is Sufficient: Single-Positive Multi-Label Learning with Label Enhancement

Authors: Ning Xu, Congyu Qiao, Jiaqi Lv, Xin Geng, Min-Ling Zhang

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on twelve corrupted MLL datasets show the effectiveness of SMILE over several existing SPMLL approaches.
Researcher Affiliation Academia 1School of Computer Science and Engineering, Southeast University, Nanjing 210096, China 2 RIKEN Center for Advanced Intelligence Project, Tokyo 103-0027, Japan {xning, qiaocy}@seu.edu.cn, is.jiaqi.lv@gmail.com, {xgeng, zhangml}@seu.edu.cn
Pseudocode Yes Algorithm 1 SMILE Algorithm
Open Source Code Yes Source code is available at https://github.com/palm-ml/smile.
Open Datasets Yes In the experiments, we adopt twelve widely-used MLL datasets [13], which cover a broad range of cases with diversified multi-label properties.
Dataset Splits Yes For each dataset, we run the comparing methods with 80%/10%/10% train/validation/test split.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or specific computational resources) used for running experiments.
Software Dependencies No The paper mentions using 'Adam optimizer [17]' but does not provide specific version numbers for any software dependencies or libraries.
Experiment Setup Yes The mini-batch size and the number of epochs are set to 16 and 25, respectively. The learning rate and weight decay are selected from {10 4, 10 3, 10 2} with a validation set.