Skip to content

Adversarial Meta Learning on Contrastive Learning model SimCLR. Aiming to robustify original SimCLR with data augmentation and adversarial training.

Notifications You must be signed in to change notification settings

Xiaoyang-Song/Robust-SimCLR-via-Adversarial-Meta-Learning

Repository files navigation

Robust-SimCLR-via-Adversarial-Meta-Learning

This project aims to improve the robustness of contrastive learning model SimCLR via adversarial learnings and enhance its generalizability across all attacks by incoporating the meta-learning framework.

Keyword: Robust Machine Learning, Meta Learning, Adversarial Learning, SimCLR, etc.

About

Adversarial Meta Learning on Contrastive Learning model SimCLR. Aiming to robustify original SimCLR with data augmentation and adversarial training.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages