VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.

263-2400-00L 6 Credits MSC , WBZ D-INFK , D-MATH , D-GESS , D-ITET

Reliable and Trustworthy Artificial Intelligence

Lecturers & Examiners: Prof. Dr. Martin Vechev
VVZ CR 4.0

Last Updated: 2026-06-03 00:07:33

Abstract

Reliability, security, privacy, and robustness are core challenges in achieving trustworthy AI and are of fundamental importance. The goal of this course is to teach both the mathematical foundations of this emerging field and to introduce students to the latest and most exciting advances.

Objective

Upon completion of the course, students should have mastered the underlying methods and be able to apply them to a variety of engineering and research problems. To facilitate deeper understanding, the course includes a group coding project where students will build a system based on the learned material.

Content

The course is structured in three parts: Robustness in Machine Learning ------------------------------------------------- - Adversarial attacks and defenses on deep learning models. - Automated certification of deep learning models (convex relaxations, branch and bound, randomized smoothing). - Certified training of deep neural networks (combining symbolic and continuous methods). - State-of-the-art attacks and novel attack vectors for large language models (LLMs). Privacy in Machine Learning ------------------------------------------------- - Threat models (e.g., data stealing, model poisoning, membership inference). - Privacy attacks in decentralized (federated) machine learning. - Protection via differential privacy; applications to centralized and decentralized model training. - Memorization in generative AI models; training data extraction attacks. - Private attribute inference with generative AI models. - Securing data flows in agentic AI systems. Provenance and Evaluation in Generative AI ------------------------------------------------- - Reliable detection of AI-generated content via watermarking. - Removing and forging watermarks; data watermarking. - Dataset contamination: detecting and evading detection. - Trustworthy evaluation of LLMs: challenges in benchmarking and rating. - Bridging AI regulation (e.g., EU AI Act) and technical evaluations. More at: https://www.sri.inf.ethz.ch/teaching/rtai25

General Information

Language
English
Levels
MSC , WBZ
Frequency
Yearly recurring

Examination

Type
session examination
Mode
written 180 minutes
Aids
Two A4-pages (i.e. one two-sided or two one-sided A4-sheets of paper), either handwritten or 11 point minimum font size.
30% of your grade is determined by mandatory project work and 70% is determined by a written exam.Students who are repeating the course are required to repeat the project work.

Course Components

Type Title Time & Place Hours
lecture Reliable and Trustworthy Artificial Intelligence No time listed 2 h weekly
exercise Reliable and Trustworthy Artificial Intelligence
Exercise session will start in the second week of the semester.
No time listed 2 h weekly
independent project Reliable and Trustworthy Artificial Intelligence No time listed 1 h weekly

Offered In