Adversarial ML & Security Threats

Secure the Model—Understand, Detect, and Defend Against Adversarial AI Attacks

MODE
Mode(Online) TYPE
Mentor Based LEVEL
Moderate

Skills you will gain:

Adversarial ML & Security Threats is an advanced, research-driven training program that explores how malicious actors exploit weaknesses in machine learning systems. As AI becomes central to decision-making in defense, finance, healthcare, and cybersecurity, understanding adversarial threats is essential. This course provides technical insights into how models can be tricked, poisoned, or reverse-engineered, and trains participants to build defenses against such attacks using robust ML practices, secure deployment methods, and adversarial training.

Aim:

To develop advanced capabilities in identifying, analyzing, and mitigating adversarial machine learning (AML) attacks and AI-specific vulnerabilities in deployed ML systems, with a focus on real-world security threats in AI-enabled environments.

Program Objectives:

To bridge machine learning engineering with cybersecurity expertise
To build capabilities for defending against real-world AI attacks
To create secure, reliable, and resilient AI systems
To train professionals for AI red teaming and adversarial simulation roles

What you will learn?

Week 1: Foundations of Adversarial Machine Learning

Module 1: Introduction to Adversarial ML

Chapter 1.1: What is Adversarial ML?
Chapter 1.2: Historical Context and Emerging Importance
Chapter 1.3: Types of Adversarial Threats (White-box, Black-box, Gray-box)
Chapter 1.4: Overview of Vulnerabilities in ML Pipelines

Module 2: Attacks Against ML Models

Chapter 2.1: Evasion Attacks on Image, Text, and Tabular Models
Chapter 2.2: Poisoning Attacks During Training
Chapter 2.3: Model Inversion and Membership Inference
Chapter 2.4: Tools and Libraries (Foolbox, ART, CleverHans)

Week 2: Defensive Strategies and Robust Model Design

Module 3: Making Models Robust

Chapter 3.1: Adversarial Training Techniques
Chapter 3.2: Input Preprocessing and Gradient Masking
Chapter 3.3: Certified Defenses and Formal Guarantees
Chapter 3.4: Evaluation Metrics for Robustness

Module 4: Security in the ML Lifecycle

Chapter 4.1: Secure Data Pipelines and Label Integrity
Chapter 4.2: Attack Surface in Model Deployment
Chapter 4.3: Threat Modeling for ML Systems
Chapter 4.4: Secure MLOps and Monitoring Pipelines

Week 3: Real-World Applications and Future Challenges

Module 5: Adversarial ML in Practice

Chapter 5.1: Case Studies: Attacks on Facial Recognition, NLP, and Healthcare Models
Chapter 5.2: Adversarial Threats in Federated Learning and Edge AI
Chapter 5.3: Legal, Ethical, and Compliance Risks
Chapter 5.4: AI Red Teaming and Offensive Testing

Module 6: Capstone and Emerging Trends

Chapter 6.1: Design Your Own Adversarial Attack Scenario
Chapter 6.2: Simulate and Evaluate Defense Mechanisms
Chapter 6.3: Final Capstone Project Presentation
Chapter 6.4: Future Directions – AI Security, Regulation, and Red-Blue Team Dynamics

Intended For :

AI/ML engineers, cybersecurity professionals, and researchers
Graduate students and advanced learners in computer science or data science
Proficiency in Python, ML frameworks (TensorFlow, PyTorch), and basic cybersecurity concepts is recommended