Metashift: A dataset of datasets for evaluating contextual distribution shifts and training conflicts

W Liang, J Zou - arXiv preprint arXiv:2202.06523, 2022 - arxiv.org
Understanding the performance of machine learning models across diverse data
distributions is critically important for reliable applications. Motivated by this, there is a …
被引用次数：62 相关文章
[PDF] nsf.gov

[PDF][PDF] MetaShift: a dataset of datasets for evaluating contextual distribution shifts and training conflicts.

W Liang, J Zou - International Conference on Learning Representations, 2022 - par.nsf.gov
Understanding the performance of machine learning models across diverse data
distributions is critically important for reliable applications. Motivated by this, there is a …

MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflicts

W Liang, J Zou - International Conference on Learning Representations - openreview.net
Understanding the performance of machine learning models across diverse data
distributions is critically important for reliable applications. Motivated by this, there is a …

MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflicts

W Liang, J Zou - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
Understanding the performance of machine learning models across diverse data
distributions is critically important for reliable applications. Motivated by this, there is a …