Bidirectional Adversarial Training for Semi-Supervised Domain Adaptation
Authors: Pin Jiang, Aming Wu, Yahong Han, Yunfeng Shao, Meiyu Qi, Bingshuai Li
IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the effectiveness of Bi AT on three benchmark datasets and experimental results demonstrate the proposed method achieves the state-of-the-art. We evaluate Bi AT on three benchmark datasets and conduct extensive ablation study. |
| Researcher Affiliation | Collaboration | Pin Jiang1,2 , Aming Wu1,2 , Yahong Han1,2 , Yunfeng Shao3 , Meiyu Qi3 and Bingshuai Li3 1College of Intelligence and Computing, Tianjin University, Tianjin, China 2Tianjin Key Lab of Machine Learning, Tianjin University, Tianjin, China 3Huawei Noah s Ark Lab {jpin, tjwam, yahong}@tju.edu.cn, {shaoyunfeng, qimeiyu, libingshuai}@huawei.com |
| Pseudocode | Yes | Algorithm 1 Bidirectional Adversarial Training (Bi AT) |
| Open Source Code | No | The paper does not provide any explicit statement about making source code available or a link to a code repository. |
| Open Datasets | Yes | Domain Net is a recent benchmark dataset for large-scale domain adaptation that has 6 domains and 345 classes [Peng et al., 2019]. Office contains 3 domains with 31 classes and we eliminate the domains with less examples and construct 2 scenarios, Webcam to Amazon (W A) and DSLR to Amazon (D A). Office-Home contains 4 domains (Real, Clipart, Art, Product) with 65 classes. |
| Dataset Splits | Yes | For each dataset, we use the randomly selected one or three labeled examples per class as the labeled target examples (1shot and 3-shot) by [Saito et al., 2019]. Other three labeled target examples are used as validation set and the remaining are used as unlabeled target data. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running the experiments. |
| Software Dependencies | No | The paper mentions using Alex Net and Res Net34 as backbones, and SGD as an optimizer, but does not provide specific version numbers for any software dependencies. |
| Experiment Setup | Yes | We set λ1 in Eq. (11) as 1, λ2 as 2. To avoid misjudgments of Le-vat in early training, we use cosine warm-up on λ3. Lmme is sensitive to datasets, even sub-domains of a same dataset, and we choose λ4 [0.1, 2.5] for different scenarios. We adopt SGD as an optimizer with initial learning rate 0.01. In order to make a fair comparison, other setup we choose the same as [Saito et al., 2019]. |