Our poster this afternoon at #ICLR2023: why calibration is not enough, and how to estimate the remainder to quantify uncertainty
Indeed, a classifier can be over-confident on some individuals and under-confident on others. How to separate this error from irreducible uncertainty?
Our paper on data augmentation for EEG data was accepted by the Journal of Neural Engineering!
It all started w the nice work done by @JP4illard during his internship @Parietal_INRIA! Congrats to him & our amazing co-authors @tomamoral & @agramfort!
Short 🧵 on our findings 👇!