# model-poisoning
標記為「model-poisoning」的 5 篇文章
Federated Learning 攻擊s
攻擊ing federated learning through model update poisoning, gradient leakage, free-rider attacks, and Byzantine fault exploitation.
federated-learninggradient-leakagemodel-poisoningbyzantinefree-rider
AI 供應鏈安全概覽
AI/ML 供應鏈攻擊面的完整概覽,涵蓋模型投毒、資料投毒、相依性攻擊與與 OWASP LLM03:2025 對齊的風險評估框架。
supply-chainowasprisk-assessmentmodel-poisoningdata-poisoningdependenciesdefense
聯邦學習安全
聯邦學習的安全威脅,包含拜占庭節點、梯度洩漏、後門注入與隱私攻擊。
infrastructurefederated-learningmodel-poisoningprivacy
Lab: Model Supply Chain Poisoning
Simulate model supply chain attacks by injecting 後門s into model weights distributed through public registries.
labssupply-chainmodel-poisoningadvanced
實作:攻擊聯邦學習
動手實作實驗,在使用 Flower 框架模擬的聯邦學習環境中實作模型投毒攻擊:拜占庭攻擊、模型替換,並量測攻擊影響。
labfederated-learninghands-onflowerbyzantinemodel-poisoningsimulation