Interpolation for Robust Learning: Data Augmentation on Wasserstein Geodesics

Jiacheng Zhu, Jielin Qiu, Aritra Guha, Zhuolin Yang, Xuanlong Nguyen, Bo Li, Ding Zhao
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:43129-43157, 2023.

Abstract

We propose to study and promote the robustness of a model as per its performance on a continuous geodesic interpolation of subpopulations, e.g., a class of samples in a classification problem. Specifically, (1) we augment the data by finding the worst-case Wasserstein barycenter on the geodesic connecting subpopulation distributions. (2) we regularize the model for smoother performance on the continuous geodesic path connecting subpopulation distributions. (3) Additionally, we provide a theoretical guarantee of robustness improvement and investigate how the geodesic location and the sample size contribute, respectively. Experimental validations of the proposed strategy on four datasets including CIFAR-100 and ImageNet, establish the efficacy of our method, e.g., our method improves the baselines’ certifiable robustness on CIFAR10 upto 7.7%, with 16.8% on empirical robustness on CIFAR-100. Our work provides a new perspective of model robustness through the lens of Wasserstein geodesic-based interpolation with a practical off-the-shelf strategy that can be combined with existing robust training methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-zhu23i, title = {Interpolation for Robust Learning: Data Augmentation on {W}asserstein Geodesics}, author = {Zhu, Jiacheng and Qiu, Jielin and Guha, Aritra and Yang, Zhuolin and Nguyen, Xuanlong and Li, Bo and Zhao, Ding}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {43129--43157}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/zhu23i/zhu23i.pdf}, url = {https://proceedings.mlr.press/v202/zhu23i.html}, abstract = {We propose to study and promote the robustness of a model as per its performance on a continuous geodesic interpolation of subpopulations, e.g., a class of samples in a classification problem. Specifically, (1) we augment the data by finding the worst-case Wasserstein barycenter on the geodesic connecting subpopulation distributions. (2) we regularize the model for smoother performance on the continuous geodesic path connecting subpopulation distributions. (3) Additionally, we provide a theoretical guarantee of robustness improvement and investigate how the geodesic location and the sample size contribute, respectively. Experimental validations of the proposed strategy on four datasets including CIFAR-100 and ImageNet, establish the efficacy of our method, e.g., our method improves the baselines’ certifiable robustness on CIFAR10 upto 7.7%, with 16.8% on empirical robustness on CIFAR-100. Our work provides a new perspective of model robustness through the lens of Wasserstein geodesic-based interpolation with a practical off-the-shelf strategy that can be combined with existing robust training methods.} }
Endnote
%0 Conference Paper %T Interpolation for Robust Learning: Data Augmentation on Wasserstein Geodesics %A Jiacheng Zhu %A Jielin Qiu %A Aritra Guha %A Zhuolin Yang %A Xuanlong Nguyen %A Bo Li %A Ding Zhao %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-zhu23i %I PMLR %P 43129--43157 %U https://proceedings.mlr.press/v202/zhu23i.html %V 202 %X We propose to study and promote the robustness of a model as per its performance on a continuous geodesic interpolation of subpopulations, e.g., a class of samples in a classification problem. Specifically, (1) we augment the data by finding the worst-case Wasserstein barycenter on the geodesic connecting subpopulation distributions. (2) we regularize the model for smoother performance on the continuous geodesic path connecting subpopulation distributions. (3) Additionally, we provide a theoretical guarantee of robustness improvement and investigate how the geodesic location and the sample size contribute, respectively. Experimental validations of the proposed strategy on four datasets including CIFAR-100 and ImageNet, establish the efficacy of our method, e.g., our method improves the baselines’ certifiable robustness on CIFAR10 upto 7.7%, with 16.8% on empirical robustness on CIFAR-100. Our work provides a new perspective of model robustness through the lens of Wasserstein geodesic-based interpolation with a practical off-the-shelf strategy that can be combined with existing robust training methods.
APA
Zhu, J., Qiu, J., Guha, A., Yang, Z., Nguyen, X., Li, B. & Zhao, D.. (2023). Interpolation for Robust Learning: Data Augmentation on Wasserstein Geodesics. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:43129-43157 Available from https://proceedings.mlr.press/v202/zhu23i.html.

Related Material