FLIP
Backdoor defense framework
A framework for defending against backdoor attacks in federated learning systems
[ICLR 2023, Best Paper Award at ECCV’22 AROW Workshop] FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning
44 stars
3 watching
2 forks
Language: Python
last commit: about 1 year ago backdoorbackdoor-attacksbackdoor-defensebyzantinecomputer-visiondefensedistributed-computingfederated-learningprivacypythonpytorchsecurity
Related projects:
| Repository | Description | Stars |
|---|---|---|
 ebagdasa/backdoor_federated_learning | An implementation of a framework for backdoors in federated learning, allowing researchers to test and analyze various attacks on distributed machine learning models. | 271 |
 hfzhang31/a3fl | A framework for attacking federated learning systems with adaptive backdoor attacks | 22 |
 ai-secure/crfl | This project presents a framework for robust federated learning against backdoor attacks. | 71 |
 ybdai7/chameleon-durable-backdoor | A federated learning system implementation that enables planting durable backdoors in global models by adapting to peer images. | 32 |
 ksreenivasan/ood_federated_learning | Researchers investigate vulnerabilities in Federated Learning systems by introducing new backdoor attacks and exploring methods to defend against them. | 64 |
 git-disl/lockdown | A backdoor defense system against attacks in federated learning algorithms used for machine learning model training on distributed datasets. | 14 |
 nil0x42/phpsploit | A tool allowing attackers to remotely execute commands and maintain persistence on compromised web servers using stealthy PHP backdoors. | 2,221 |
 dcalab-unipv/turning-privacy-preserving-mechanisms-against-federated-learning | This project presents an attack on federated learning systems to compromise their privacy-preserving mechanisms. | 8 |
 zlijingtao/ressfl | Develops techniques to improve the resistance of split learning in federated learning against model inversion attacks | 20 |
 jhcknzzm/federated-learning-backdoor | An implementation of a federated learning attack method known as Neurotoxin, which introduces backdoors into machine learning models during the training process. | 63 |
 ethz-spylab/rlhf_trojan_competition | Detecting backdoors in language models to prevent malicious AI usage | 107 |
 ganyuwang/vfl-czofo | A unified framework for improving privacy and reducing communication overhead in distributed machine learning models. | 11 |
| ai-secure/fedgame | An implementation of a game-theoretic defense against backdoor attacks in federated learning. | 5 |
 jeremy313/soteria | An implementation of a defense against model inversion attacks in federated learning | 55 |
 tobiabocchi/flipperzero-bruteforce | Automates brute-force attacks on fixed OOK codes used in SubGHz protocols | 2,011 |