Unsupervised Domain Adaptation Based on Source-Guided Discrepancy

Authors: Seiichi Kuroki, Nontawat Charoenphakdee, Han Bao, Junya Honda, Issei Sato, Masashi Sugiyama4122-4129

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, experimental results demonstrate the advantages of S-disc over the existing discrepancy measures. [...] We demonstrate the effectiveness of S-disc for unsupervised domain adaptation in experiments.
Researcher Affiliation Collaboration The University of Tokyo, Tokyo, Japan RIKEN, Tokyo, Japan
Pseudocode Yes Algorithm 1: S-disc Estimation for the 0-1 Loss
Open Source Code No The paper cites "ar Xiv:1809.03839" as a longer version of the paper, but does not provide a direct link to source code for the described methodology. It mentions using "scikit-learn" and "picos/cvxopt" as third-party tools but not its own implementation code.
Open Datasets Yes Here, we used MNIST (Le Cun, Cortes, and Burges 2010) dataset [...] Clean source domains: Five grayscale MNIST-M (Ganin et al. 2016), [...] Target domain: MNIST.
Dataset Splits No The paper specifies the total number of examples used for source and target domains (e.g., 'The number of examples was ranged from {1000, 2000, ..., 20000} for each domain'), but it does not provide explicit train/validation/test split percentages, absolute counts for each split, or references to predefined splits for reproduction.
Hardware Specification Yes The simulation was run on 2.8GHz Intel R Core i7.
Software Dependencies No The paper mentions using 'scikit-learn (Pedregosa et al. 2011)', 'picos' and 'cvxopt', but it does not specify version numbers for these software dependencies, which are crucial for reproducibility.
Experiment Setup Yes logistic regression implemented with scikit-learn with default parameters (Pedregosa et al. 2011). [...] The Gaussian noise with standard deviation ϵ = 30, 40, 50 were added and clipped to force the value to between 0 255.