Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Multi-Label Open Set Recognition
Authors: Yibo Wang, Jun-Yi Hang, Min-Ling Zhang
NeurIPS 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on various datasets validate the effectiveness of the proposed approach in dealing with the MLOSR problem. 4 Experiments |
| Researcher Affiliation | Academia | Yi-Bo Wang, Jun-Yi Hang, Min-Ling Zhang School of Computer Science and Engineering, Southeast University, Nanjing 210096, China Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, China EMAIL |
| Pseudocode | Yes | Algorithm 1 The pseudo-code of SLAN |
| Open Source Code | Yes | Our source code can be found at https://palm.seu.edu.cn/zhangml/. |
| Open Datasets | Yes | Table 1 summarizes the detailed characteristics of each benchmark multi-label data set S employed in the experiments, including the number of instances |S|, number of features dim(S), number of class labels L(S), label cardinality LCard(S), label density LDen(S), number of distinct label sets DL(S) and proportion of distinct label sets PDL(S). The datasets used in this paper are public, and can be found in Section 4.1. |
| Dataset Splits | No | The paper states: 'we sample 60% instances without unknown labels to form training set while the remaining instances are treated as test data.' It does not explicitly mention a separate validation set or split for hyperparameter tuning during training. |
| Hardware Specification | Yes | A Linux server equipped with Intel Xeon CPU (48 cores @ 2.67GHz) and 256GB memory is used for supporting the experiments. |
| Software Dependencies | No | The paper does not explicitly list specific software dependencies with version numbers (e.g., Python version, library versions like PyTorch, TensorFlow, or scikit-learn with their specific version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | For the proposed SLAN approach, trade-off parameters are set as α = 0.1, β = 0.1, γ = 10, µ1 = 0.1, τ = 0.8. µ0 is fixed to be 0.1. |