Probabilistic Artificial Intelligence (2021)

How can we build systems that perform well in uncertain environments and unforeseen situations? How can we develop systems that exhibit “intelligent” behavior, without prescribing explicit rules? How can we build systems that learn from experience in order to improve their performance? We will study core modeling techniques and algorithms from statistics, optimization, planning, and control and study applications in areas such as sensor networks, robotics, and the Internet. The course is designed for upper-level undergraduate and graduate students. VVZ information is available here.
News
  • [21/09/2021] Due to COVID-19 regulations at ETH, only students with Covid certificates ( COVID Certificate app, valid ETH CovMass test, foreign vaccination certificates accepted by ETH ) and ETH student card are allowed to enter the classroom, either for lectures or tutorials. Please bring them to class. 
  • [20/09/2021] The first project (Task 0) has already been released, which will be open for submissions until the end of this semester (December 19th, 2021). This project is intended as a playground for you to get familiar with the project platform, and thus is ungraded.
  • The first lecture starts on 24.09, and the first tutorial starts on 30.09.
  • Some resources on this site and links to streams may be protected with custom passwords. To obtain all passwords (including for the Q&A on Zoom) click here when you are within the ETH network. To be within the ETH network, you need to be either physically present at ETH to use the ETH Wi-Fi or you must establish a VPN connection to ETH.
  • Welcome the the course Probabilistic Artificial Intelligence!

Lectures
Lectures will be held on Friday, 10 am to 12 pm and 13 am to 14 am, in ETA F5 with a simultaneous video screening in ETF E1. A link to a live-stream will also be made available and allow following the lectures from home on your personal device. If you follow the live-stream, you will be able to ask questions via the course channel on the EduApp . The lecture will be recorded and the recording will be made available after the lecture here . The first lecture will take place on Friday, 24th September 2021, and the last lecture will take place on Friday, 17th December 2021. It is not mandatory to attend the lectures. In order to attend the lectures physically, you require a Covid-19 certificate per ETH policy. Unfortunately, even with the certificate, ETH requires you to wear a mask during the lecture. Links below will become active when the resource becomes available.
Date Topic Slides Annotated Slides Recording
Fri 24.09. Introduction 01-introduction.pdf 01-introduction-annotated.pdf part1 part2
Fri 01.10. Bayesian Linear Regression empty empty empty
Fri 08.10. Gaussian Processes empty empty empty
Fri 15.10. Gaussian Processes II empty empty empty
Fri 22.10. Variational Inference empty empty empty
Fri 29.10. Markov Chain Monte Carlo empty empty empty
Fri 05.11. Bayesian Deep Learning empty empty empty
Fri 12.11. Active Learning empty empty empty
Fri 19.11. Markov Decision Processes empty empty empty
Fri 26.11. Reinforcement Learning empty empty empty
Fri 03.12. Reinforcement Learning II empty empty empty
Fri 10.12. Reinforcement Learning III empty empty empty
Fri 17.12. Model-based Deep RL empty empty empty

Tutorials
Tutorials will be held on Thursday, 4 pm to 6 pm in CHN C14. A link to a live-stream will also be made available and allow following the tutorials from home on your personal device. If you follow the live-stream, you will be able to ask questions via the course channel on the EduApp . The tutorial will be recorded and the recording will be made available after the tutorial here . The first tutorial will take place on Thursday, 30th September 2021, and the last tutorial will take place on Thursday, 16th December 2021. It is not mandatory to attend the tutorials. In order to attend the tutorials physically, you require a Covid-19 certificate per ETH policy. Unfortunately, even with the certificate, ETH requires you to wear a mask during the tutorial. Links below will become active when the resource becomes available.
Date Topic Slides Recording Homework/ Solution Moodle
Thu 30.09. Math/ Probability Tutorial 1 Tutorial 1 Homework 1 Moodle

Q&A sessions
Q&A sessions (virtual office hours) will be held on Monday, 5 pm to 6 pm virtually on Zoom. The Q&A sessions are an informal opportunity to ask questions about the course. We may use some Q&A sessions to give students the opportunity to present their project work. The Q&A session will be recorded and the recording will be made available after the Q&A session. You will be able to ask questions via the native Zoom chat or by speaking out, if requested to do so. The first Q&A session will take place on Monday, 27th September 2021, and the last Q&A session will take place on Monday, 13th December 2021. It is not mandatory to attend the Q&A sessions.
Date Topic Recording
Mon 27.10. Session I Recording

Contact
Instructor Prof. Andreas Krause
Head TA Max Paulus
Assistants Michael Aerni, Andisheh Amrollahi, Junting Chen, Sebastian Curi, Salmane El Messoussi, Ya-Ping Hsieh, Mohammad Reza Karimi, Parnian Kassraie, David Lindner, Anastasia Makarova, Marco Milanta, Mojmír Mutný, Seyedmorteza Sadat, Ramesh Shyam, Carl Johann Simon Gabriel, Vignesh Somnath, Paul Streli, Xinyu Sun, Scott Sussex, Lenart Treven
Mailing List Please use Moodle for any questions regarding the course or ask your question in the lectures, tutorials or Q&A sessions. If you need to contact the Head TA or the lecturer directly, please send an email to pai21-info@inf.ethz.ch. Please think twice before you send an email though and make sure you read all information here carefully.
Lectures
Fri 10-12 ETA F5 [ETF E1] Stream
Fri 13-14 ETA F5 [ETF E1] Stream
Tutorials
Thu 16-18 CHN C14 Stream
Questions & Answers
Mon 17-18 Virtual Zoom

Moodle
We do not maintain a mailing list, but instead kindly request you to use Moodle to ask questions with regard to the course. Please ask your questions in the Moodle forum whose topic best fits your question. If you need to contact the Head TA directly, please send an email to pai21-info@inf.ethz.ch instead of contacting me at my personal email address. I will not respond to requests sent to my personal email address. Please think twice before you send an email though and make sure you read all information here carefully. Based on previous experience, we received a lot of questions or requests that are resolvable with the information provided here. We previously used Piazza instead of Moodle to moderate questions. We are trialling Moodle for the first time this semester. If you have previously used Piazza and would like to provide feedback regarding the user experience on Moodle, we appreciate you sending your thoughts to pai21-moodlefeedback@inf.ethz.ch . We may not directly reply, but we will carefully evaluate your feedback.
Projects
The course includes a total of five projects. Projects are code assignments that require solving machine learning problems with methods taught in the course. For each project, you are allowed to work in a group of one to three students. It is your own responsibility to form a group and you can find teammates in the lectures or on Moodle. The first project (Task 0) will be ungraded; its purpose is to help you familiarize yourself with the code submission workflow. The remaining projects are graded and are accounted for in determining your final grade for the course. More information including a tentative schedule is available in the project information sheet and on the project server . Both are accessible from within the ETH network or via VPN.
Homework
We will publish a total of six (optional) homework assignments during the lecture series. The homework assignments will be published on this website and some questions from the homework assignment will additionally be made available as a Moodle quiz . These assignments are intended for you to apply and reinforce the material presented in the lecture and to get accustomed to the Moodle platform. You are encouraged, but not required to do the homework. Doing the homework assignments is not mandatory. Your performance in the homework will have no influence on your final grade. Homework assignments are expected to be published bi-weekly, with solutions following one week after or being directly visible in Moodle. The exact day and time a homework is being published may deviate slightly over the course of the semester.
Demos
Demos will be shown during the lecture and are made available to you here. They are hosted in a GitLab repository to which you need to be given access. Everyone who enrolled (on mystudies) to the course before Monday, 20th September 2021, will be automatically granted access to this Gitlab repository by Thursday, 23rd September 2021. If you enrolled at a later date, please individually request access by sending an email to Junting Chen , only after 23/09/2021. Use the subject line “Access Request: PAI 2021 Demos” and include your nethz in this email. The demos are Jupyter Notebooks.
Exam
The exam will last for 120 minutes and might be computer-based (Moodle). The language of examination is English. More information about the exam will be communicated nearer towards the examination session
Performance Assessment
In order to pass this course and be allowed to sit the session examination, you need to pass the projects. This is, you need to achieve a project grade of 4 or higher. If you don’t pass the projects, you are required to de-register from the exam and will otherwise be treated as a no-show. The final grade is computed as a weighted average of the session exam grade (70%) and the project grade (30%). There are no special arrangements for PhD students who are taking this course. In order to obtain a “Testat”, you need a passing grade for the course. This is you need to pass the projects as described above and achieve an overall passing grade (4 or higher). If you passed the projects last year, you still need to do the projects again this year. The project grade cannot be carried over from the previous year.
Text Books
  • S. Russell, P. Norvig. Artificial Intelligence: A Modern Approach (4th edition).
  • C. E. Rasmussen, C. K. I. Williams Gaussian Processes for Machine Learning.
  • Christopher M. Bishop. Pattern Recognition and Machine Learning. [optional]
  • Richard S. Sutton and Andrew G. Barto. Reinforcement Learning: An Introduction.