arXiv Machine Learning
Fine-grained Distributed Backdoor Attacks in Federated Learning
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Sep 24
Backdoors Leave Structural Traces: FedMAST for Backdoor Detection and Containment in Federated Learning
The paper introduces FedMAST, a Federated Multi‑Axis Structural Tracing defense designed to detect and contain backdoor attacks in federated learning. FedMAST evaluates client updates through complementary structural, spectral, and historical evidence, applying tiered filtering and round‑level containment. In experiments across six backdoor attacks, FedMAST consistently achieves lower attack success rates while preserving high main‑task accuracy.
arXiv AI
Sep 10
Unveiling Hidden Threats: Using Fractal Triggers to Boost Stealthiness of Distributed Backdoor Attacks in Federated Learning
The paper introduces Fractal-Triggered Distributed Backdoor Attack (FTDBA), a new method that uses fractal self‑similarity to strengthen sub‑triggers in federated learning backdoor attacks. By employing a dynamic angular perturbation mechanism, FTDBA reduces the amount of poisoned data needed while maintaining a 92.3% attack success rate. Experiments show a 22.8% lower detection rate and a 41.2% reduction in KL divergence compared to traditional distributed backdoor attacks.
arXiv Machine Learning
Sep 23
FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated Learning
FedNIA is a defense framework for federated learning that identifies and excludes malicious clients without needing a central test dataset. It works by injecting random noise inputs and analyzing layerwise activation patterns with an autoencoder to detect abnormal behaviors caused by data poisoning. The method can counter various attack types—including sample poisoning, label flipping, and backdoors—even when multiple attackers collaborate, and shows strong performance on non‑iid federated datasets.
arXiv Computer Vision
Aug 28
Checkerboard: Closed-Form and Data-Independent Trigger Design for Clean-Label Backdoor Attacks
The paper introduces Checkerboard, a clean‑label backdoor attack that uses a closed‑form, data‑independent trigger design based on an input‑space Fisher‑separability objective and a ridge four‑neighbor local‑smoothness prior. This approach yields a pixel‑wise checkerboard trigger without requiring data access, surrogate model training, or iterative optimization, and it outperforms existing norm‑bounded clean‑label attacks across four benchmark datasets. On CIFAR‑10, poisoning 20 samples with a 10/255 perturbation achieves a 95.72% attack success rate, while on IN‑100 a 0.4% global poisoning rate yields over 83% ASR without harming clean accuracy, and the attack remains robust against state‑of‑the‑art defenses.
arXiv AI
Aug 10
TOFD: Target-Oriented Feature Decoupling against Poisoning Attacks in Split Federated Learning
arXiv:2608. 07274v1 Announce Type: cross Abstract: Split Federated Learning (SFL) facilitates privacy-preserving collaborative training with reduced client-side overhead.
arXiv Machine Learning
Jun 25
Color Matters: Trigger Color Affects Success in Federated Backdoor Attacks
arXiv:2606. 25858v1 Announce Type: cross Abstract: Federated learning is vulnerable to backdoor attacks in which malicious clients inject poisoned updates while preserving benign-task performance.