Theme 3 - Large Language Models
Lecture 7 - Adversarial Attacks on LLMs
Adversarial Machine Learning
Course Information
Previous Offerings
Syllabus
Course Description
Textbooks
Learning Outcomes
Prerequisites
Grading
Course Schedule
Lecture 1 - Introduction to AML
Theme 1 - Foundations
Lecture 2 - Deep Learning Overview
Lecture 3 - Mathematics for Machine Learning
Theme 2 - Evasion Attacks
Lecture 4 - White-box Evasion Attacks
Lecture 5 - Black-box Evasion Attacks
Theme 3 - Large Language Models
Lecture 6 - Introduction to LLMs
Lecture 7 - Adversarial Attacks on LLMs
Theme 4 - Defenses and Poisoning Attacks
Lecture 8 - Defenses Against Evasion Attacks
Lecture 9 - Poisoning Attacks and Defenses
Theme 5 - AML in Cybersecurity
Lecture 10 - Network Intrusion Detection
Lecture 11 - Malware Detection and Classification
Lecture 12 - Spam Filtering, URL Detection, Cyber-physical Systems
Theme 6 - Privacy, Explainability, Fairness
Lecture 13 - Privacy Attacks and Defenses
Lecture 14 - Explainability in Machine Learning
Lecture 15 - Bias and Fairness in Machine Learning
Code Examples
Overview
FGSM Attack in TensorFlow
White-box Evasion Attacks in TensorFlow
White-box Evasion Attacks in PyTorch
Black-box Boundary Attack
LLM Supervised Fine-tuning
Jailbreak Attacks on LLMs
Adversarial Training
Attack on Network Intrusion Detection
Hello World of ML
Overview (PDF)
Logistic Regression on Iris
Random Forest on Breast Cancer
XGBoost on Titanic
MLPs on MNIST
ResNet on CIFAR-10
DistilBERT on IMDb
Llama 2 on Alpaca
Qwen3 GRPO on MATH-500
VLM Reasoning and Tool Use on MATH-500
Theme 3 - Large Language Models
Lecture 7 - Adversarial Attacks on LLMs
Lecture 7 - Adversarial Attacks on Large Language Models
Click to Open the Lecture in a New Window
|
Download the PowerPoint Slides
Lecture 6 - Introduction to LLMs
Lecture 8 - Defenses Against Evasion Attacks