Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Unsupervised Domain Adaptation With Distribution Matching Machines
Authors: Yue Cao, Mingsheng Long, Jianmin Wang
AAAI 2018 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Comprehensive experiments validate that the DMM approach significantly outperforms competitive methods on standard domain adaptation benchmarks. We perform extensive experiments to evaluate DMM against state of the art methods on standard domain adaptation benchmarks including both image and text datasets. |
| Researcher Affiliation | Academia | Yue Cao, Mingsheng Long, Jianmin Wang KLiss, MOE; NEL-BDS; TNList; School of Software, Tsinghua University, China EMAIL EMAIL EMAIL |
| Pseudocode | No | The paper has a section titled 'Algorithm' but it describes the optimization paradigm and steps in paragraph form and mathematical equations, not as a structured pseudocode block or a clearly labeled algorithm figure. |
| Open Source Code | No | Codes, datasets and configurations will be made available online. |
| Open Datasets | Yes | Datasets Office-31 (Saenko et al. 2010) is the standard benchmark for domain adaptation... Caltech-256 (Griffin, Holub, and Perona 2007) is a standard database... In experiments, we adopt Office-10 + Caltech-10 (Fernando et al. 2013; Long et al. 2013)... Reuters-21578 is a text dataset of Reuters news articles. |
| Dataset Splits | No | We use all source examples with labels and all target examples without labels for training, and report the average classification accuracy. For all comparison methods, we select their optimal hyper-parameters by cross-validation on labeled source data as (Pan et al. 2011). |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU/CPU models, memory specifications, or cloud resources) used to run the experiments. |
| Software Dependencies | No | The paper mentions using the 'LIBSVM package' and 'quadprog in MATLAB' for solving optimization problems, but it does not specify any version numbers for these software components or any other libraries. |
| Experiment Setup | Yes | We give parameter sensitivity analysis for DMM, which will validate that DMM can achieve stable performance for a wide range of hyper-parameter settings. We check sensitivity of subspace dimension r and penalty parameter λ. Figures 3(b) and 3(c) show DMM outperforms baselines for wide ranges of parameters r [10, 70], λ [10 6, 10 4]. |