awesome-aml
github.com/wangjksjtu/awesome-aml ↗A curated list of awesome adversarial attack and defense papers
46
GitHub Stars
74
Curated Resources
2
Categories
3 hours ago
Last Refreshed
AttackDefense
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me modifying the training schemes or models resources from awesome-aml"
Installation instructions →What's inside
Defense
- Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural NetworkModifying the training schemes or models
- Adversarial Logit PairingModifying the training schemes or models
- Adversarially Robust Generalization Just Requires More Unlabeled DataModifying the training schemes or models
- Are Labels Required for Improving Adversarial Robustness?Modifying the training schemes or models
- A study of the effect of JPG compression on adversarial imagesModifying the adversraial examples
- Beyond Adversarial Training: Min-Max Optimization in Adversarial Attack and DefenseModifying the training schemes or models
Attack
- Adversarial examples in the physical worldWhite-Box (Gradient-based)
- Adversarial Objects Against LiDAR-Based Autonomous Driving SystemsRobust physical attack
- Adversarial Risk and the Dangers of Evaluating Against Weak AttacksBlack-Box (Gradient-free)
- AutoZOOM: Autoencoder-based Zeroth Order Optimization Method for Attacking Black-box Neural NetworksBlack-Box (Gradient-free)
- BayesOpt Adversarial AttackBlack-Box (Gradient-free)
- Black-box Adversarial Attacks with Bayesian OptimizationBlack-Box (Gradient-free)
Showing a sample of 74 resources. View the full list on GitHub →