Hey - I think this one is strong https://vitaldb.net/dataset/
What are the best datasets to benchmark out-of-distribution generalization? E.g. EEG/MEG recordings, etc.
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Perhaps it shouldn't be a dataset. It can be a data generation process - algorithmic or physical black box.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
stocks
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What does it mean to be "the best dataset?"
-
Something that supports your priors while frustrating your intuitions? But what does that mean? Why couldn't you make an artificial dataset that simulates this... or if you could then maybe... why would you?
End of conversation
New conversation -
-
-
TUH-EEG which is the bigger EEG dataset in the world.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Perhaps you can use any dataset and assume that some classes are not available during training.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
From my experience there are so many internal (brain plasticity) and environmental factors that affect the recording session of EEG signals that you will even have a hard time to generalize between rec. sessions from different days.
-
There was an issue with a series of papers a while back that was caused by this inter-session generalizability (the lack of it). A paper that was explaining this and was invalidating the previous ones was published:https://arxiv.org/abs/1812.07697
End of conversation
New conversation -
-
-
Physionet bank
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.