= Project Machine Learning (Projekt Maschinelles Lernen) =

== General Information ==

||'''First meeting'''||Wednesday, 18 October 2017, 10:15am, Room MAR 4.065 ||
||'''Responsible'''|| [[http://www.ml.tu-berlin.de/menue/mitglieder/klaus-robert_mueller/|Prof. Dr. Klaus-Robert Müller]] ||
||'''Contact'''|| Jacob Kauffmann jacob.r.kauffmann@campus.tu-berlin.de ||
||'''Language'''|| Englisch ||
||'''Credit'''|| M.Sc. Modul (Projekt), 9 LP (ECTS), PÄS ||
||'''Course Information'''|| [[https://isis.tu-berlin.de/course/view.php?id=11366]] ||

== First Meeting ==

Our first meeting is Wed 18 Oct 2017 at 10:15am in Room MAR 4.065.

== Enrollment / Limited number of participants ==

The course has a a limit of 10 participants. If you intend to participate, please send an e-mail to jacob.r.kauffmann@campus.tu-berlin.de as early as possible since we assign spots mostly on a first come / first serve basis. It is mandatory to attend the first meeting or e-mail the course organizer if this should not be possible. Otherwise we assume that you have no interest in the course anymore and give your spot to someone else.

== Description ==

/* All information can be found on ISIS: [[https://isis.tu-berlin.de/course/view.php?id=8141]]. */ A [[https://tubcloud.tu-berlin.de/s/mtGrOyXzztb8v0f/download|PDF overview]] is available. The course has an [[http://www.lsf.tu-berlin.de/qisserver/servlet/de.his.servlet.RequestDispatcherServlet?state=verpublish&status=init&vmfile=no&publishid=198639&moduleCall=webInfo&publishConfFile=webInfo&publishSubDir=veranstaltung|entry in Vorlesungsverzeichnis]].

=== Prerequisites ===

There are no formal entry requirements. However, we strongly recommend
 * the lecture ''Machine Learning 1'',
 * the course ''Python Programming for Machine Learning'' and
 * ideally, the ''Lab Course Machine Learning''.

Students who do not have attended these lecture, should make sure they have the following skills:
 * '''Basic theory:''' Students should be fluent in probability theory, linear algebra and understand how and why mainstream learning algorithms work.
 * '''Some practical ML experience:''' Prior exposure to the practical application of ML algorithms is essential; students should know how to select hyperparameters and assess the performance of a trained predictor.
 * '''Python programming:''' All code that is to be handed in must be written in Python; students should be able to program in Python using the packages `numpy` and `scipy`.