Eugene Bagdasarian

Backdoors Framework for Deep Learning and Federated Learning. A light-weight tool to conduct your research on backdoors.

ebagdasa

Source code for paper "How to Backdoor Federated Learning" (https://arxiv.org/abs/1807.00459)

ebagdasa

Code for "Differential Privacy Has Disparate Impact on Model Accuracy" NeurIPS'19

ebagdasa

Salvaging Federated Learning by Local Adaptation

ebagdasa

Code for paper: "Spinning Language Models: Risks of Propaganda-as-a-Service and Countermeasures"

ebagdasa

simple Differential Privacy in PyTorch

ebagdasa

ebagdasa