"Safe and Pleasant ML": Today 4:30-6pm, then we go for the weekly Walk in the Park

[Artur Dubrawski]

Team,

Today we will have a "ML equalizer" session in which I will talk about some
basic concepts of survival in the Machine Learning research and
applications world. It will be primarily targeting our largest-than-ever
cohort of summer interns, who come from various backgrounds to work with us
at the Lab on various ML or ML-related projects.

But everyone is welcome. In particular if you are a relatively new member
of our team and in the process of immersing yourself into the ways of
thinking about ML and AI, you could find attending this meeting quite
useful.

Topics I am thinking about covering today:
* Basic ML tasks
* Challenge of overfitting and how to mitigate the risk of it
* Challenges caused by random data and how to mitigate them
* Empirical evaluation of classifiers: ROC, McNemar's test

Let me know if you think we should adjust that menu or if we should cover
more topics.
We may arrange more sessions like this over the summer, and I will be happy
to share the floor with willing volunteer speakers as well.

This will be a hybrid meeting in NSH 3305 and on zoom (Meeting ID:
94267966836 Passcode: 960534).
I plan to join in person.

We expect to be done by or before 6pm, right in time for the walk in the
park.

For those of us who would like to participate in the walk, please note that
this time we will be meeting in front of Newell-Simon Hall at 6pm, instead
of our usual starting spot.

Cheers
Artur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20220624/d2856f4b/attachment.html>