NDSS2022

In the blog, I summary the accepted papers from NDSS 2022 which are related to my research interests from the link.

Paper list:

• Session 2C: ML and AI
  • Tetrad: Actively Secure 4PC for Secure Training and Inference
  • MIRROR: Model Inversion for Deep Learning Network with High Fidelity
  • Local and Central Differential Privacy for Robustness and Privacy in Federated Learning
    • The paper evaluate the effect of LDP and CDP to protect the FL model robustness from membership, backdoor attacks.
  • DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection
    • Analyze the data distribution features to identify malicious model updates in the federated learning
• Session 4C: ML and AI
  • What You See is Not What the Network Infers: Detecting Adversarial Examples Based on Semantic Contradiction
    • Analyze the adversarial examples’ internal features.
    • A pretty interesting story.
  • Euler: Detecting Network Lateral Movement via Scalable Temporal Graph Link Prediction
    • Temporal GNN model.
  • Fooling the Eyes of Autonomous Vehicles: Robust Physical Adversarial Examples Against Traffic Sign Recognition Systems
    • Adversarial attack on the objective detection models like YOLO v5.
  • FedCRI: Federated Mobile Cyber-Risk Intelligence
    • Cyber-Threat Intelligence
• Session 5C: Attacks on ML and AI
  • ATTEQ-NN: Attention-based QoE-aware Evasive Backdoor Attacks
    • Attention-based evasive backdoor attack.
    • The paper is interesting: Leveraging the image generator techniques to generate the trigger or backdoor!
  • RamBoAttack: A Robust and Query Efficient Deep Neural Network Decision Exploit
    • A query efficient attack.
  • Property Inference Attacks Against GANs
  • Get a Model! Model Hijacking Attack Against Machine Learning Models

DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection

paper/code

Main Idea

Key insight

Experiments