Model-Based Domain Generalization

Authors: Alexander Robey, George J. Pappas, Hamed Hassani

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In our experiments, we report improvements of up to 30% over state-of-the-art domain generalization baselines on several benchmarks including Colored MNIST, Camelyon17-WILDS, FMo W-WILDS, and PACS.
Researcher Affiliation Academia Department of Electrical and Systems Engineering University of Pennsylvania {arobey1,pappasg,hassani}@seas.upenn.edu
Pseudocode Yes Algorithm 1 Model-Based Domain Generalization (MBDG)
Open Source Code Yes Our code is publicly available at the following link: https://github.com/arobey1/mbdg.
Open Datasets Yes In our experiments, we report improvements of up to 30% over state-of-the-art domain generalization baselines on several benchmarks including Colored MNIST, Camelyon17-WILDS, FMo W-WILDS, and PACS.
Dataset Splits Yes Model selection for each of these datasets was performed using hold-one-out cross-validation. For Camelyon17-WILDS and FMo W-WILDS, we use the repository provided with the WILDS dataset suite, and we perform model-selection using the out-of-distribution validation set provided in the WILDS repository.
Hardware Specification No The paper states "Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes]" in the ethics checklist, but the provided text does not contain specific hardware details like GPU/CPU models or memory.
Software Dependencies No The paper does not explicitly state specific software dependencies with version numbers.
Experiment Setup Yes Further details concerning hyperparameter tuning and model selection are deferred to Appendix E. (3.b) Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes]