Safe Multi-View Deep Classification

Authors: Wei Liu, Yufei Chen, Xiaodong Yue, Changqing Zhang, Shaorong Xie

AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experiments on various kinds of multi-view datasets validate that the proposed SMDC method can achieve precise and safe classification results. In this section, we extensively evaluate the proposed method on real-world multi-view datasets and compare it with existing multi-view classification methods.
Researcher Affiliation Collaboration 1 College of Electronics and Information Engineering, Tongji University, Shanghai, China 2 School of Computer Engineering and Science, Shanghai University, Shanghai, China 3 Artificial Intelligence Institute of Shanghai University, Shanghai, China 4 College of Intelligence and Computing, Tianjin University, Tianjin, China 5 VLN Lab, NAVI Med Tech Co., Ltd. Shanghai, China
Pseudocode Yes Algorithm 1: Algorithm for Safe Multi-View Deep Classification (SMDC)
Open Source Code No The paper does not provide any explicit statements about the release of source code or links to a code repository for the described methodology.
Open Datasets Yes We conduct experiments on six real-world multi-view datasets as follows: Handwritten (Van Breukelen et al. 1998), Scene15 (Fei Fei and Perona 2005), Animal (Lampert, Nickisch, and Harmeling 2013), Caltech101 (Fei-Fei, Fergus, and Perona 2004), CUB (Wah et al. 2011) and HMDB (Kuehne et al. 2011).
Dataset Splits Yes We then use 5-fold cross-validation to select the learning rate from 1e 4, 3e 4, 1e 3, 3e 3 . For all datasets, 20% samples are used as test sets.
Hardware Specification Yes The model is implemented by Py Torch on one NVIDIA A100 with GPU of 40GB memory.
Software Dependencies No The paper mentions 'The model is implemented by Py Torch', but it does not specify a version number for PyTorch or any other software dependencies, which is required for reproducibility.
Experiment Setup Yes The Adam optimizer (Kingma and Ba 2014) is used to train the network, where l2-norm regularization is set to 1e 5. We then use 5-fold cross-validation to select the learning rate from 1e 4, 3e 4, 1e 3, 3e 3 .