Adversarial Machine Learning Course

Learn how to perform white-box and black-box attacks on machine learning classifiers in this course.

1 hour, 33 minutes

Course description

This course begins by providing an overview of white-box and black-box adversarial attacks on machine learning systems. It will then guide you through using the Fast Gradient Signed Method (FGSM) white-box attack on a keras machine learning model. Next, we will cover black-box attacks. You will be guided on using a machine learning as a service system called Clarif.AI and then performing a black-box adversarial attack to trick this service into labeling a benign image as dangerous. Finally, to solidify learning, the student is given an assignment on tricking a MNIST keras classifier via a white-box adversarial attack.


Assignment: TrickMe

File - 01:00:00

Assignment: TrickMe video

Video - 00:01:00

Black-Box Attack on Clarif.AI

Video - 00:21:00

Code from scratch and perform a black-box attack on Clarif.AI’s moderation model.
Getting Started with Clarif.AI

Video - 00:03:00

Tutorial on the offerings of Clarif.AI ML as a service, installation and setup of the developer API key.
White-Box Attacks on Machine Learning

Video - 00:06:00

Understand and perform the FGSM attack on a keras ResNet model using the Foolbox library.
Adversarial Machine Learning

Video - 00:02:00

Overview of adversarial attacks on ML.

Meet the author

Emmanuel Tsukerman

Dr. Tsukerman graduated from Stanford University and UC Berkeley. In 2017, his machine-learning-based anti-ransomware product won Top 10 Ransomware Products by PC Magazine. In 2018, he designed a machine-learning-based malware detection system for Palo Alto Network's WildFire service (over 30,000 customers).

In 2019, Dr. Tsukerman authored the Machine Learning for Cybersecurity Cookbook and launched the Infosec Skills Cybersecurity Data Science Learning Path.

Unlock 7 days of free training

  • 1,400+ hands-on courses and labs
  • Certification practice exams
  • Skill assessments

Associated NICE Work Roles

All Infosec training maps directly to the NICE Workforce Framework for Cybersecurity to guide you from beginner to expert across 52 Work Roles.

  • Law Enforcement / Counterintelligence Forensics Analyst
  • Cyber Defense Forensics Analyst
  • Data Analyst

Plans & pricing

Infosec Skills Personal

$299 / year

  • 190+ role-guided learning paths (e.g., Ethical Hacking, Threat Hunting)
  • 100s of hands-on labs in cloud-hosted cyber ranges
  • Custom certification practice exams (e.g., CISSP, Security+)
  • Skill assessments
  • Infosec peer community support

Infosec Skills Teams

$799 per license / year

  • Team administration and reporting
  • Dedicated client success manager
  • Single sign-on (SSO)
    Easily authenticate and manage your learners by connecting to any identity provider that supports the SAML 2.0 standard.
  • Integrations via API
    Retrieve training performance and engagement metrics and integrate learner data into your existing LMS or HRS.
  • 190+ role-guided learning paths and assessments (e.g., Incident Response)
  • 100s of hands-on labs in cloud-hosted cyber ranges
  • Create and assign custom learning paths
  • Custom certification practice exams (e.g., CISSP, CISA)
  • Optional upgrade: Guarantee team certification with live boot camps

Learn about scholarships and financing with

Affirm logo

Award-winning training you can trust