News
13.05.2020
|
GradesDear students,
the grades of the write-up and projects as well as the final grade are available on your Personal Status page. If you want feedback for your write up, please feel free to contact Verena Wolf and ask for an appointment. We hope you all... Read more Dear students,
the grades of the write-up and projects as well as the final grade are available on your Personal Status page. If you want feedback for your write up, please feel free to contact Verena Wolf and ask for an appointment. We hope you all enjoyed the seminar as much as we did. We wish you all the best for your future (studies).
the R4L-team |
07.03.2020
|
Download ReviewsDear students, you reviews are now available on your Personal Status page.
Best, the R4L-team
|
26.02.2020
|
Review TemplateDear students, as some of you requested a template for the review, you can now find it unter materials. Best, the R4L-team |
21.02.2020
|
Upload Write-upDear students, due to technical reasons some of you we're unable to upload your write-up on your personal status page. The function should now be available for everyone. If there are further problems, just send the slides by mail to Timo. Best, the... Read more Dear students, due to technical reasons some of you we're unable to upload your write-up on your personal status page. The function should now be available for everyone. If there are further problems, just send the slides by mail to Timo. Best, the R4L-team |
05.02.2020
|
Computation Time for MCTS AlgorithmsDear students, for MCTS-algorithms (including all variants of Connect4Zero) we will give you 1 second of computation time per step. the R4L-team |
09.12.2019
|
Results TalksDear students, you can now find the results of your talk on your personal status page. If you want personal feedback, you can get it in the Office Hour. the R4L-team |
05.12.2019
|
Presentations TomorrowDear students, we are looking forward to your presentations tomorrow! |
25.11.2019
|
Reminder Deadline Slides, Git Registration, Group NameDear students, this is just a friendly reminder that your slides are due this Friday. Please send them by email to your supervisor on Friday. Furthermore, there are two more tasks to be done until December 4: Dear students, this is just a friendly reminder that your slides are due this Friday. Please send them by email to your supervisor on Friday. Furthermore, there are two more tasks to be done until December 4: We are looking forward to your presentations and hope you are doing fine. Best, the team of R4L |
25.10.2019
|
Room ChangeDear students, due to technical reasons we move the tutorial to E1.3 room 401. |
23.10.2019
|
Neural Network TutorialDear students, the tutorial will take place on Friday 10 am in room SR 015 in E1.3.
Dear students, the tutorial will take place on Friday 10 am in room SR 015 in E1.3.
There won't be time to install the programs within the tutorial, so please make sure in advance that everything is working. Best, the team of R4L |
18.10.2019
|
Topic SelectionDear students, we hope you liked todays meeting and are now highly motivated for the upcoming seminar. You can now find all the provided sources for your talks under Papers & Sources. The slides will be published later today under Materials. We want to... Read more Dear students, we hope you liked todays meeting and are now highly motivated for the upcoming seminar. You can now find all the provided sources for your talks under Papers & Sources. The slides will be published later today under Materials. We want to point out, that you are not only supposed to read the papers of your category (a,b &c), but all papers of your topic(1,2,3 or 4), such that you can discuss the topics together. Only for the presentation (and the write-up) the separation into categories is needed. Please keep this in mind when voting for or against a topic. Best, the team of R4L |
23.09.2019
|
Kick-off MeetingDear students, the Kick-off will take place on October 18 at 10 am in E1.3 Room 328. Participation is absolutely mandatory if you want to attend the seminar. Best, the team of Rein4cement Learning
|
16.09.2019
|
RegistrationDear students, due to a large number of applicants and because we want to assign the seminar slots as fair as possible, we moved the final registration to https://seminars.cs.uni-saarland.de/seminars1920. Please register there except if you study bioinformatics... Read more Dear students, due to a large number of applicants and because we want to assign the seminar slots as fair as possible, we moved the final registration to https://seminars.cs.uni-saarland.de/seminars1920. Please register there except if you study bioinformatics as the bioinformatics seminars do not use this system. Slots for bioinformatics students will be assigned the traditional way. Best, the team of Rein4cement Learning |
11.07.2019
|
Pre-registrationThe Pre-registration is available from now until the 15th of September. Please only register, if you seriously plan to attend the seminar. As the number of participants is limited, we will select from the pre-registered students. The pre-registration does not... Read more The Pre-registration is available from now until the 15th of September. Please only register, if you seriously plan to attend the seminar. As the number of participants is limited, we will select from the pre-registered students. The pre-registration does not guarantee participation in the seminar. |
Rein4cement Learning
Reinforcement Learning (RL) is a popular subdiscipline of Machine Learning for problems that require strategies to solve complex tasks such as board games, scheduling problems or other discrete optimization problems.
This seminar will discuss the theoretical basis of RL as well as popular RL-algorithms such as Deep Q-Learning. In the final phase of the seminar, the participants will apply these algorithms to the well-known game Connect 4.
The seminar will include short presentations by the participants, a programming project that will be solved in small groups as well as a write-up.
The Kick-off will take place on October 18 at 10 am in E1.3 Room 328. Participation is absolutely mandatory if you want to attend the seminar.
Literature:
[1] Richard S. Sutton and Andrew G. Barto: Reinforcement Learning - An Introduction (Second Edition)
[2] David Silver, Julian Schrittwieser, Karen Simonyan et al. : Mastering the game of Go without human knowledge
[3] Ahmad Hammoudeh : A Concise Introduction to Reinforcement Learning