Hi all, so, I think it won't be necessary to have the individual sessions if we instead have individual sessions after the assignment is due, where we go over the tasks and their solutions with each team in person. However, if this is not planned (due to covid) and the individual meetings should still be online, I think it might make sense to still have this. Nevertheless, a help-desk-like slot where everyone with questions can just come and ask individually (not during the lecture) could also be helpful. Best, Bastian Am 03.03.22 um 12:28 schrieb Arno Feiden:
Hi,
last year I did one additional session where I met with the students in their respective teams for 10-15 minutes each, checking up on their progress after a week of their two week assignments. I gave them individual feedback on their finished assignment "one way" as an email.
I liked those individual session, because this way I could interact with everyone not just the three outgoing people that dared to ask questions during and after the lecture.
I also think it is a good idea to plan some spare time for the second lecture to address problems with the assignments that concern everyone.
Kind regards Arno
----- Original Message ----- From: "Jochen Garcke" <garcke@ins.uni-bonn.de> To: "mllab" <mllab@ins.uni-bonn.de> Sent: Thursday, 3 March, 2022 11:57:09 Subject: Re: [Mllab] Teilnehmerzahl
Hi,
remind me, how did we do the assistance last time, was it one or two times ? How do we want to do it this time ? Since we have the lectures on-site, you could allow some time after the second lecture ? And some online appointments on top of that ?
Regards, Jochen
On 03/03/2022 11:28, Jochen Garcke wrote:
Hi,
Linear Least Squares and Nearest Neighbors - Jochen 1st lecture, Paolo second lecture, Paolo assistance for the students and evaluation SVM - Jochen the lectures, Ivan assistance for the students and evaluation PCA - Sara Deep Learning - Ivan Reinforcement Learning - Arno
Grüße, Jochen
On 01/03/2022 11:49, Jochen Garcke wrote:
Hi,
wir wechseln ja. Wir könnten dann noch einen Slot erlauben um auf glatte 20 zu kommen ? Bzw. eine Warteliste, falls jemand absagt.
Plan ist PCA - Sara Deep Learning - Ivan Reinforcement Learning - Arno
Noch final zu entscheiden, wo und was Paolo mitmacht.
Linear Least Squares and Nearest Neighbors SVM
Bastian, du wärst dann als 'Berater' involviert ?
Grüße, Jochen
On 01/03/2022 10:52, Bastian Bohn wrote:
Hi,
bei 2er Gruppen hätten wir ja dann ca. 9-10 Gruppen. Das ist schon ordentlich Aufwand, wenn eine einzige Person korrigiert und bewertet, aber wenn wir pro Zettel wechseln, sollte das gehen, denke ich.
Viele Grüße, Bastian
Am 01.03.22 um 10:47 schrieb Jochen Garcke:
Hallo zusammen,
wir haben dann doch einige Anmeldung Bachelor 10 Master 9 2 die mitmachen wollen, aber keine Prüfung.
Ich weiß jetzt nicht, was wir als Obergrenze gesehen haben ?
Im SS19 sehe 21 ? Im WS18/19 3 + 22, wobei wir dann dreier-Gruppen hatten.
Sollte also gehen mit zweier Gruppen ? Oder 3er für Bachelor ?
Grüße, Jochen
-- Dr. Bastian Bohn Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI Abteilung Numerische Datenbasierte Vorhersage (NDV) Schloss Birlinghoven 53757 Sankt Augustin Tel.: +49 2241 14 4225 https://www.scai.fraunhofer.de/de/geschaeftsfelder/numerische-datenbasierte-...