Class Overwhelms: Mutual Conditional Blended-Target Domain Adaptation

Authors: Pengcheng Xu, Boyu Wang, Charles Ling

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments. We evaluate our method based on standard BTDA tasks... We summarize the standard BTDA in Table 1. Ablation and Analysis.
Researcher Affiliation Academia 1 Western University, London, ON N6A 5B7, Canada 2 Vector Institute, Toronto, ON M5G 1M1, Canada
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper states 'We followed the implementations in (Junguang Jiang, Baixu Chen, Bo Fu, Mingsheng Long 2020)' which refers to baseline implementations, but does not provide a concrete access link or explicit statement for the source code of their own methodology.
Open Datasets Yes Datasets. We evaluate our method based on standard BTDA tasks (Chen et al. 2019; Roy et al. 2021a): Office-31 (Saenko et al. 2010), Office-Home (Venkateswara et al. 2017), Domain Net (Peng et al. 2019), and a specialized dataset Office Home-LMT for label shift in BTDA.
Dataset Splits No The paper states 'For evaluation, we use one domain as the source and the rest as blended targets' but does not provide specific training, validation, or test dataset splits.
Hardware Specification No The paper does not provide specific hardware details such as GPU or CPU models used for running its experiments.
Software Dependencies No The paper mentions 'SGD optimizer' but does not provide specific ancillary software details with version numbers.
Experiment Setup Yes For all datasets, we use SGD optimizer with learning rates η0 = 0.01, α = 10, and β = 0.75. We set the uncertainty threshold γ = 0.05 for all datasets. Since CST and SENTRY use Auto Augment for data augmentation, we set the number of transformations N = 1 and the transform severity M = 2.0 in Auto Augment (Lim et al. 2019) for a fair comparison.