Professional Documents
Culture Documents
AdversarialML GiovIEIIT 20230413
AdversarialML GiovIEIIT 20230413
Online – 13/04/2023
1 2 3
Training data ML model Prediction
poisoning corruption failure
Considered attacks:
• Carlini-Wagner (CW)
• Jacobian based Saliency MAps (JSMA)
• Fast Gradient Sign Method (FGSM)
Thursday Seminars - IEIIT Youth, 13/04/2023
CARLINI WAGNER
Most relevant
features
Value ranking
Combination of
the most relevant
intervals
Thresholds tuning
Adversarial region
• Start with the legitimate class (y = 0) • Start with the attack class (y = 1)
feature/value ranking feature/value ranking
SVDD + LLM
JSMA - DNS
FGSM - RUL
Thursday Seminars - IEIIT Youth, 13/04/2023
RESULTS: XAI-BASED RELIABLE AI
LLM0%
Visual LLM rules comparison for the three adversarial attacks (DNS tunneling dataset)
zeroFPRSVDD explainableSVDD
Thursday Seminars - IEIIT Youth, 13/04/2023
MEMORY CONSUMPTION
• Open data:
https://www.kaggle.com/datasets/cnrieiit/adversarial-machine-learning-dataset
Q&A