When Source-Free Domain Adaptation Meets Learning with Noisy Labels

Authors: Li Yi, Gezheng Xu, Pengcheng Xu, Jiaqi Li, Ruizhi Pu, Charles Ling, Ian McLeod, Boyu Wang

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments demonstrate significant improvements to existing SFDA algorithms by leveraging ETP to address the label noise in SFDA.
Researcher Affiliation Academia 1Department of Statistical and Actuarial Sciences 2Department of Computer Science University of Western Ontario
Pseudocode Yes Algorithm 1: SFDA ELR Source Free Domain Adaptation with ELR
Open Source Code No The paper does not provide an explicit statement about releasing source code or a link to a code repository.
Open Datasets Yes We use four benchmark datasets... Office-31 (Saenko et al., 2010), Office-Home (Venkateswara et al., 2017), Vis DA (Peng et al., 2017) and Domain Net (Peng et al., 2019).
Dataset Splits No The paper mentions using benchmark datasets but does not explicitly provide specific train/validation/test split percentages or sample counts, nor does it refer to predefined splits with citations for reproducibility.
Hardware Specification No The paper mentions ResNet architectures used as backbones but provides no specific details on the hardware (GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper discusses various methods and loss functions but does not specify software dependencies (e.g., programming languages, libraries, or frameworks) with version numbers.
Experiment Setup Yes We set the learning rate to 1e-4 for all layers except for the last two FC layers, where we apply 1e-3 for the learning rate for all datasets. The hyperparameter β is chosen from {0.5, 0.6, 0.7, 0.8, 0.9, 0.99}, and λ is chosen from {1, 3, 7, 12, 25}. Table 5: Optimal Hypermaraters (β/λ) on various datasets.