Elements of Machine Learning Prof. Dr. Isabel Valera, Prof. Dr. Jilles Vreeken

News

26.11.2021

Help tutorials next week exclusively online

Dear students,

In light of the current Corona situation, next week's help tutorials will take place exclusively online. We will increase the number of tutors helping you to ensure that all questions are answered in the best possible way.

We look forward to... Read more

Dear students,

In light of the current Corona situation, next week's help tutorials will take place exclusively online. We will increase the number of tutors helping you to ensure that all questions are answered in the best possible way.

We look forward to seeing you online next week!

Stay safe,

Your EML Team

25.11.2021

Checklist for your next submission

Dear all,

I hope you are all well and healthy. We would like to provide you with a checklist for your next assignment submission:

For the new assignment, I have...

  • created (formally) a new team on CMS. This can be with my previous partner, if I like.... Read more

Dear all,

I hope you are all well and healthy. We would like to provide you with a checklist for your next assignment submission:

For the new assignment, I have...

  • created (formally) a new team on CMS. This can be with my previous partner, if I like. Important is, that I created a new team on CMS.  
  • filled out the code of conduct for the new assignment, including inserting the current assignment number, and updated the date. I signed the code of conduct. I submitted it individually in my own CMS, even, if I am a team (each team member submits their own code of conduct).
  • put my name on each of the pages I submitted. If I am in a team, I have put BOTH team members' names on each page that I submit.
  • written on plain paper with blue or black ink/font color. I have not used checkered paper.
  • numbered the equations in my proofs. This allows the tutor to reference while grading and provides me with more precise feedback.
  • referred to my equations by the numbers I gave them, if I needed to explained my equations. I used underbraces to refer to parts of the equation. I did not write "RHS of this equation", because it adds ambiguity.
  • written question numbers next to my code in R (as a comment). This allows the tutor to understand my code better and faster.

Looking forward to seeing you at the next help tutorial, where you can get help for your next submission and ask clarification questions regarding the feedback you received in your last assignment. If you have questions on the submission guidelines or style, kindly ask in the forum. We also constantly update the FAQ - please check them before sending us an email.

Stay healthy,

Your EML Team

22.11.2021

Points Assignment 1 released

Dear Students,

We have just released points for Assignment 1. Our tutors have done an amazing job not only checking the assignments, but also providing you with feedback. Please check it out. If you remain with questions, please use the help tutorials to ask for... Read more

Dear Students,

We have just released points for Assignment 1. Our tutors have done an amazing job not only checking the assignments, but also providing you with feedback. Please check it out. If you remain with questions, please use the help tutorials to ask for clarification of your grading.

Please note, if you have not submitted a Code of Conduct, - as announced - you have received 0 points.

For a few students, we have not yet released points. That means, we are still checking your submission. We kindly ask for your patience and will notify you.

 

Best regards,

Your EML team

 

15.11.2021

Problems with installing package glmnet

Dear students,

If you have problems installing the glmnet package you can try the following:

 

1. Install RTools4.0 
https://cran.r-project.org/bin/windows/Rtools/
and try again to install glmnet with,

install.packages("glmnet")

2. If it still... Read more

Dear students,

If you have problems installing the glmnet package you can try the following:

 

1. Install RTools4.0 
https://cran.r-project.org/bin/windows/Rtools/
and try again to install glmnet with,

install.packages("glmnet")

2. If it still does not work, download the binaries form 
https://cran.r-project.org/web/packages/glmnet/index.html 
and install glmnet with the command 

install.packages("/path/to/glmnet_4.1-3.zip", repos=NULL)

Make sure to run RStudio (or R) as administrator otherwise it might cause permission issues  when compiling packages. 

 

Please do not contact the professors nor the tutors directly, your emails will not be forwarded. Read CMS, the forum, ask on the forum or send us an email to eml-ta [at] lists.saarland-informatics-campus.de

Please do not contact us via email regarding your feedback to the previous assignment. Please attend  the help tutorials and ask there.

 

FAQ

1. Oganization of the course: Please read this information carefully. A pre-recorded lecture (Lecture 0) with all organizational details is available for you (see News). If you remain with questions afterwards, kindly join us in the first tutorial, where you will be able to ask questions.

2. Coronavirus: Please find here the university FAQ for attending lectures in person: https://www.uni-saarland.de/en/page/coronavirus/faq.html All participants in a face-to-face event at the university (in teaching and non-teaching) must provide proof of 3G according to the current provincial regulation on the Corona pandemic, i.e. either complete vaccination, recovery or a negative test (twice per working week). Proof is provided by using and truthfully stating in the Staysio app; for this purpose, all premises used for attendance events must be mapped in the Staysio app. Alternatively, you can register via the web application (English version) described here in the lower part. Otherwise, a German Apple/Google account can also be used for the download.Should participants not be able to register themselves via smartphone, they are obliged to submit their contact details and 3G proof to the event management immediately at the beginning of the event using the following form (data protection declaration). If the form is used, it must be collected by the lecturers, kept in accordance with data protection regulations and destroyed after a period of four weeks.

3. I took EML last semester and I passed all the assignment sheets - do I need to do the assignments again this year, to be able to register for the exam? Yes for reasons of fairness, you will need to do them again.

4. Please abstain from contacting the professors directly. If you have questions, check this page, check the forum, post your question in the forum and - last resort - contact us via  eml-ta [at] lists.saarland-informatics-campus.de

5. Practical submissions only in R. As clearly explained on both this website and in Lecture 0 (Organization) we will only consider practical assignment submissions in R.

6. What is the difference between the EML (this course) and ML (last summer semester) course?  ML is more advanced, comprehensive and comes with more mathematical depth than EML. It is likely that ML be offered again in Summer 2022, but not yet decided. So unfortunately we cannot yet guarantee this at the moment.

7. Can I team up with a partner who is not in the same tutorial as me?  Yes.

8. How is the final grade evaluated?  100% exam.

9. Does every team member have to hand in their own signed code of conduct?  Yes.

10. Do I have to hand in signed a code of conduct for every assignment? Yes.

11. Can I sign the code of conduct digitally? Yes.

12. What happens, if I do not hand in the code of conduct? We will not grade submissions for which there is no code of conduct submitted.

13. Do you publish the solutions to the assignments? No. We only present them in the solution tutorial.

14. Do you record solution tutorials? No.

15. What do I do, if I wish a clarification on the feedback/grading I got in previous assignments? Please attend the help tutorials and ask there. Unfortunately we are not able to give you individual feedback via e-mail. Do not send us an individual email, we cannot answer them.

16. How should my submission look like? Please check out the checklist in the news section.

17. Do I need to create a new team on CMS for each assignment? Yes, for each assignment you need to create a new team. However, this can still be with your previous partner. Just set it up new on CMS.

 

For more FAQ please check the Forum.

 

Elements of Machine Learning

In this course, we will discuss the foundations—the elements—of machine learning. In particular, we will focus on the ability of, given a data set, to choose an appropriate method for analysing it, to select the appropriate parameters for the model generated by that method and to assess the quality of the resulting model. Both theoretical and practical aspects will be covered. What we cover will be relevant for computer scientists in general, as well as for other scientists involved in data analysis and modelling.

This course replaces the course Elements of Statistical Learning, and will be held in English.

 

Prerequisites: The course is targeted to students in computer science, bioinformatics, maths, and general sciences with a mathematical background. Students should know linear algebra and have good basic knowledge of statistics, for example by having taken Mathematics for Computer Scientists I and II (for linear algebra) and Statistics Lab or Mathematics for Computer Scientists III (for statistics).

 

Important resources to start the course:

  • Introduction tutorial to R. Make sure to check this pre-recorded tutorials, recorded by Osman Ali Mian, in order to get a nice and short introduction to the R language that you will need for the tutorials during the course.
  • Self-test. You can use this exercise sheet to evaluate whether you have the required background to attempt EML. For the coding part you will need to also download the ozone.Rdata dataset.
  • Lecture 0 - Organizational details. Recording by Prof. Vreeken on the organizational details of the course. Make sure to check the video and the slides, which you can find in the Material section on CMS.

 

Registration

Students for which EML is a required part of their Bachelor programme can take it as a Basic Lecture, while all other students (Bachelor or Master) can take it as an Advanced Lecture. There is no difference between the two, except for how it will be listed on your transcript, either way you receive 6 ECTS for successfully completing the exam. Obviously, you can only get credits for the course once.

 

Contact

We have prepared a forum for you to discuss your questions and find study groups during the course. You can find it on the top bar on CMS.

Please before reaching out to us,

1. Check, if the information is available on this CMS website, in the lecture material, or in the recommended reading.

2. Check, if the information is available in trusted online sources and check with your fellow students.

3. Check the discussion forum, if your peers have already asked the question.

4. Post your question in the online forum using the correct category and subject line. This allows your peers to learn from your questions.

Only in exceptional cases, where your question is private/personal, please email us to: eml-ta  [at] lists [dot] saarland-informatics-campus [dot] de. Please do not send us private message via the forum. We will not be able to answer them.

 

Organizational Information

Lectures

Lectures will start in the week of October 18th!

In the lectures we will cover the main theoretical aspects of the course.

Lectures*:  Thursdays from 16:15h to 17:45h (Gebäude E2 2 - Hörsaal 0.01 (Günter-Hotz-Hörsaal)) Link for Zoom: See Information > How to access lecture/tutorials?

Lectures will also be accessible online. More information will follow.

 

Tutorials

During Solution tutorials, the TAs and tutors will present the correct solution and answer questions about them. These tutorials are only held online. During the Help tutorials, TAs and tutors give individual and group-wise assistance. These tutorials will be held both online and in-person on campus.

Please make sure that you submit a filled-out copy of the code of conduct (you can find in the material section) for each assignment.

Tutorials*: Group A: Mondays 12:15-13:45h, Group B: Tuesdays 12:15-13:45h. Location: See Information > How to access lecture/tutorials?

You will get assigned to one of the groups based on your preference that you can indicate, when you register for the course. Please attend only the assigned tutorial.

Privacy disclaimer: We have decided to use Zoom for the tutorials, as it provides superior functionality and usability for lecturing, including seamless live interaction and smooth integration of a whiteboard. We thus encourage you to join the Zoom with your real name, your camera on, and ask questions verbally. However, this is of course voluntary. If you are concerned about privacy, we encourage you to enter the Zoom meeting under a nickname or pseudonym, and use only the textual chat for communication.

 

Evaluation

The course will be evaluated in a final exam (which will most likely take place in written form). To be eligible to participate in the exams, you will need to have cumulatively scored 50% of the points for the theoretical exercises and 50% of the points for the programming exercises. To participate in the exams, you will need to register at least one week before the exam via the LSF/HISPOS system of Saarland University.

We will inform you, when the exam registration on HISPOS/LSF opens. If you are not able to register for the exam in HISPOS/LSF, please send us an email with your full name, matriculation number and study program to: eml - ta [at] lists [dot] saarland-informatics-campus [dot] de

 

Schedules

Please navigate to: Timetable for a detailed schedule of the lectures and tutorials. Please navigate to: Information > How to access the lectures? to find the respective rooms and links.

 

Lectures*:  Thursdays from 16:15h to 18:15h (Gebäude E2 2 - Hörsaal 0.01 (Günter-Hotz-Hörsaal) & oline )

Tentative Lecture Schedule
Oct 21 Lecture 1: Statistical Learning
  28 Lecture 2: Linear Regression I
Nov 4 Lecture 3: Linear Regression II
  11 Lecture 4: Classification I
  18 Lecture 5: Classification II
  25 Lecture 6: Resampling & Model Selection
Dec 2 Lecture 7: Regularization
  9 Lecture 8: Dimensionality Reduction
  16 Lecture 9: Unsupervised, Embeddings
Jan 6 Lecture 10: Clustering (online)
  13 Lecture 11: Beyond Linearity
  20 Lecture 11: Tree-based Methods
  27 Lecture 12: Support Vector Machines
Feb 3 Lecture 13: Neural Networks
  10 Lecture 14: Wrap-up with Q&A

 

 

Tutorials*: Group A: Mondays 12:15-13:45h, Group B: Tuesdays 12:15-13:45h. (Help hybrid, Solution online)

Tentative Tutorial Schedule
Assignment Type Tutorial A Tutorial B
  Intro (online!) 25.10.21 26.10.21
1 Help (hybrid) 03.11.21 (online) 02.11.21
1 Help (hybrid) 08.11.21 09.11.21
1 Solution (online) 15.11.21 16.11.21
2 Help (hybrid) 22.11.21 23.11.21
2 Help (online) 29.11.21 (online) 30.11.21 (online)
2 Solution (online) 06.12.21 07.12.21
3 Help (hybrid) 13.12.21 14.12.21
3 Help (hybrid) 10.01.22 11.01.22
3 Solution (online) 17.01.22 18.01.22
4 Help (hybrid) 24.01.22 25.01.22
4 Help (hybrid) 31.01.22 01.02.22
4 Solution (online) 07.02.22 08.02.22

 

 

Assignment Deadlines
Assignment Published Hand-in (14:00h CET)
1 October 21, 2021 November 11, 2021
2 November 11, 2021 December 02, 2021
3 December 02, 2021 January 13, 2022
4 January 13, 2022 February 03, 2022

Note: Deadlines to hand in are 14:00h CET.

 

R resources

R (version 3.2.3) is installed on the CIP pool computers and can be started by invoking R from the command line.

The official website of the R project is r-project.org. You can download R for Windows, Linux and Mac from there. Additional packages, documentation and tutorials are also available for download from the official website. Useful manuals and tutorials include:

  • R for Beginners by Emmanuel Paradis. Especially relevant for us are chapters 1, 2, 3 and 6.
  • An Introduction to R - the standard R introduction. This is a very detailed manual; it is therefore quite lengthy.

The CRAN Contributed Documentation lists many other tutorials for R beginners and advanced programmers.

You can also check out RStudio, an open-source IDE for R.

 

Bibliography

The course will, by and large, follow the book "An Introduction to Statistical Learning with Applications in R" [1]. At times, the course will take additional material from the book "The Elements of Statistical Learning" [2]. The former book is the more introductory text, the latter book is more advanced. Both books are available as free PDFs. We strongly encourage you, though, to acquire at least the first book in print. If you need to brush up on statistics, we recommend "All of Statistics" [3]. All books, as well as further background literature, are available via the library in a so-called Semesteraparat.

[1] James, W., Hastie, T. & Tibshirani, R. An Introduction to Statistical Learning with Applications in R. Springer, 2013.
[2] Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning. Springer, 2009.
[3] Wasserman, L. All of Statistics. Springer, 2005.

 

For selected lectures, we will identify interesting optional reading, such as relevant recent research papers. These we will make available here.

[4] van der Maaten, L. & Hinton, G. Visualizing data using t-SNE. Journal of Machine Learning Research, 9:2579-2605, 2008.

 

Acknowledgements

EML is based on The Elements of Statistical Learning as developed by Thomas Lengauer. We thank him for kindly sharing both materials and experience.



Privacy Policy | Legal Notice
If you encounter technical problems, please contact the administrators