spratm-asleaf/robustness-specificity

Online supplementary materials of the paper "Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity"

@Author: Dr. Xinke Li (Email: [email protected]),

@Author: Dr. Haowei Wang (Email: [email protected])

@Affiliate: Department of Industrial Systems Engineering and Management, National University of Singapore

@Date: 4 June 2024

Python Version:

Robustness-Specificity: Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity

Online supplementary materials of the paper titled

Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity

By Shixiong Wang, Haowei Wang, Xinke Li, and Jean Honorio

From the Department of Industrial Systems Engineering and Management, National University of Singapore, Singapore; School of Computing and Information Systems, University of Melbourne, Australia

Codes

• The sub-project "Deep Learning Model: BDR Learning" is maintained by Dr. Xinke Li

• The sub-project "Linear Model: BDR Support Vector Machine" is maintained by Dr. Haowei Wang

Warrant

Files/codes here are allowed to be edited, distributed, and re-used for any academic/teaching purpose without any warrant. However, you are strongly suggested to share your codes with the public if you are planning to use codes here. Let's work together to guarantee the reproducibility of experiments and the verifiability of claims in publications. We believe that this is meaningful to facilitate future research of the machine learning community.

Disclaimer

Note that the mentioned reproducibility and verifiability do not necessarily guarantee the (absolute) correctness of academic claims in a scientific publication. Future research may deny or modify or improve the philosophies, methods, models, and/or claims conveyed in this article. But readers should not try to "find bones from an egg", and the codes here are just for their reference, not for their unfriendly criticism. Of course, the authors are open to learning, and friendly comments are always welcomed.