MIT 6.S191 (2022): Deep Learning New Frontiers

[@mahinvasavada2992]

It is a privilege to be able to learn form experts like Ava and Alexander. Thank you!

[@pangxiao8983]

This series of course, is so comfortable while still being able to teach so much stuff. Amazing!

[@Lolleka]

Class starts at 7:38

[@andreas.karatzas]

I love the idea of project rewards! I would really like it if you organized a broader competition with a wide variety of DL subjects. I think it would bring together a lot of DL enthusiasts. Great lecture too :)

[@DrewGillen]

I really like the ideas behind AutoML and Graph CNNs. Great video!

[@mishrr]

Where can we listen to guest lectures. Those temple are in India.

[@porrasbrand]

I did not understand the label randomization example at 17:00 - So they trained an image/label set with (random) shuffled image / labels. So many images had as input a random label. When they confronted the trained model with a test set (correct image/label pairs) - the error was big. But when they tested it with the training set they got a high accuracy (correct labels for given input images). I assume this time the used pairs from the training set were not shuffled, so they tested the model but with correct samples of the same training set (right image/label pairs). Did I get this right? Is there more reading material on this experiment? Thanks

[@amitvyas7905]

I am MS student at CSU, Cleveland. Can I also participate in project?

[@yanhairen7293]

great lectures. Thank you so much

[@eddygonzalez6513]

Hey have a question, do you use linux, mac or windows ?

[@samirzerrouki3153]

I think linux is better and mac OS is based on Unix which is almost similar to linux ...

[@erwannmillon4609]

Thank you !!

[@ahmedb2559]

Thank you !

[@samzhao638]

One question, why did it do that in Go?

[@amir4207]

perfect

[@DeltonMyalil]

👍✔

[@Babymeosh]