Trusted Unified Feature-Neighborhood Dynamics for Multi-View Classification
Authors: Haojian Huang, Chuanyu Qin, Zhe Liu, Kaijing Ma, Jin Chen, Han Fang, Chao Ban, Hao Sun, Zhongjiang He
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on benchmark datasets show that our method improves accuracy and robustness over existing approaches, particularly in scenarios with high uncertainty and conflicting views. Extensive experiments on benchmark datasets demonstrate that our method consistently outperforms state-of-the-art approaches, achieving superior accuracy and robustness in MVC tasks, particularly in scenarios with high levels of uncertainty and conflicts. Quantative results, ablation studies, and qualitative results including visualization and robustness comparisons are presented. |
| Researcher Affiliation | Collaboration | Haojian Huang1, 2, Chuanyu Qin3, Zhe Liu4, Kaijing Ma1, 5, Jin Chen1, 5, Han Fang1, Chao Ban1, Hao Sun1*, Zhongjiang He1* 1Tele AI, China 2The University of Hong Kong, Hong Kong, China 3Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China 4Universiti Sains Malaysia, Penang, Malaysia 5Xi an Jiaotong University, Xi an, China |
| Pseudocode | No | The paper describes the methodology using text and mathematical equations, but there are no structured pseudocode or algorithm blocks explicitly labeled as such. |
| Open Source Code | Yes | Code https://github.com/Jethro James/TUNED |
| Open Datasets | Yes | Datasets. The overview of the datasets used in our study is illustrated in Table 1, with more detailed descriptions provided in the Supplementary Material. In summary, we benchmarked our approach against state-of-the-art (SOTA) methods on eight multi-view datasets. Table 1 lists PIE, Hand Written, Scene15, Caltech101, Animal, ALOI, CUB, NUS-WIDE-OBJECT. |
| Dataset Splits | No | The paper mentions using benchmark datasets (Table 1) and that 'Each method was run 10 times, and we report the average performance along with the standard deviation.' However, it does not explicitly provide specific details regarding the training, validation, or test dataset splits within the main body of the paper. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as GPU or CPU models, memory specifications, or cloud computing environments. |
| Software Dependencies | No | The paper does not specify any software dependencies or their version numbers (e.g., programming languages, libraries, or frameworks with specific versions) that would be needed to replicate the experimental environment. |
| Experiment Setup | No | The paper describes the loss functions and general training approach, including an annealing coefficient, but does not provide specific details for experimental setup such as hyperparameter values (e.g., learning rate, batch size, number of epochs, specific optimizer settings), model initialization, or specific values for loss weights like λs and λt. |