Deep Robust Unsupervised Multi-Modal Network
Authors: Yang Yang, Yi-Feng Wu, De-Chuan Zhan, Zhi-Bin Liu, Yuan Jiang5652-5659
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we propose a novel Deep Robust Unsupervised Multi-modal Network structure (DRUMN) for solving this real problem within a unified framework. The proposed DRUMN can utilize the extrinsic heterogeneous information from unlabeled data against the insufficiency caused by the incompleteness. On the other hand, the inconsistent anomaly issue is solved with an adaptive weighted estimation, rather than adjusting the complex thresholds. As DRUMN can extract the discriminative feature representations for each modality, experiments on real-world multimodal datasets successfully validate the effectiveness of our proposed method. |
| Researcher Affiliation | Collaboration | Yang Yang,1 Yi-Feng Wu,1 De-Chuan Zhan,1 Zhi-Bin Liu,2 Yuan Jiang1 1National Key Laboratory for Novel Software Technology, Nanjing University, 2Tecent {yangy, wuyf, zhandc, jiangy}@lamda.nju.edu.cn, lewiszbliu@tencent.com |
| Pseudocode | Yes | Algorithm 1 The pseudo code of DRUMN |
| Open Source Code | No | The paper does not provide any links to open-source code or explicitly state that the code for the described methodology is publicly available. |
| Open Datasets | Yes | In particular, we experiment on 4 public real-world datasets, i.e., FLICKR25K, IAPR TC-12, WIKI and NUS-WIDE, and 1 real-world incomplete multi-modal dataset, i.e., WKG Game-Hub. ... FLICKR25K: (Huiskes and Lew 2008) ... IAPR TC-12: (Escalante et al. 2010) ... WIKI: (Rasiwasia et al. 2010) ... NUS-WIDE: (Chua et al. 2009) |
| Dataset Splits | No | The paper states 'For each dataset, we randomly select 20% data for the test (query) set and the remaining instances are used for training.' which describes training and test splits, but it does not explicitly mention a separate validation set or split for hyperparameter tuning or early stopping. |
| Hardware Specification | Yes | We run the following experiments with the implementation of an environment on NVIDIA K80 GPUs server, and our model can be trained about 290 images per second with a single K80 GPGPU. |
| Software Dependencies | No | The paper mentions that 'The deep network for image encoder is implemented the same as Resnet50 (He et al. 2015)' which is a model architecture, but it does not provide specific version numbers for any software dependencies, libraries, or frameworks used in the implementation. |
| Experiment Setup | Yes | The parameter λ in the training phase is tuned in {0.1, 0.2, , 0.9}. When the variation between the objective value of Eq. 6 is less than 10 4 between iterations, we consider DRUMN converges. |