Exploiting Local Feature Patterns for Unsupervised Domain Adaptation
Authors: Jun Wen, Risheng Liu, Nenggan Zheng, Qian Zheng, Zhefeng Gong, Junsong Yuan5401-5408
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Comparisons to the state-of-the-art unsupervised domain adaptation methods on two popular benchmark datasets demonstrate the superiority of our approach and its effectiveness on alleviating negative transfer. Exhaustive experimental results on standard domain adaptation benchmarks demonstrate the promises of the proposed method by outperforming the state-of-the-art approaches. |
| Researcher Affiliation | Academia | 1Qiushi Academy for Advanced Studies, Zhejiang University, Hangzhou, Zhejiang, China 2College of Computer Science and Techology, Zhejiang University, Hangzhou, Zhejiang, China 3International School of Information Science & Engineering, Dalian University of Technology, Liaoning, China 4Department of Neurobiology, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China 5State University of New York at Buffalo |
| Pseudocode | No | The paper describes the methodology using text and mathematical equations, but does not include any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper states 'We implement our model in Tensorflow' but does not provide any concrete access information (e.g., repository link, explicit release statement) for its source code. |
| Open Datasets | Yes | We experiment on the popular Office-31 dataset (Saenko et al. 2010) and the recently introduced Office-home dataset (Venkateswara et al. 2017). |
| Dataset Splits | No | We follow standard evaluation protocols for unsupervised domain adaptation: using all labeled source data and all unlabeled target data. |
| Hardware Specification | No | Only finetuning and adapting the last two convolutional layers of VGG16 help to prevent overfitting to small datasets, reduce GPU memory footprint, and enable faster training. |
| Software Dependencies | No | We implement our model in Tensorflow and train it using Adam optimizer. |
| Experiment Setup | Yes | We keep the number of local feature patterns fixed to be 32. For local feature aggregation, we use a large α = 5000.0. We use a small similarity decay αs = 0.005 and a small sparsity threshold m = 0.02. ... we use a dropout of 0.5 over it to avoid over-fitting. ... minimize the source classification loss with a learning rate of 0.01. ... finetune the classifier, local feature patterns, and the last two convolutional layers with a learning rate of 0.0001 ... We set hyperparameters λh = 0.2, λl = 0.1 and λs = 0.1. |