I’ve been thinking we need more methods that are wrappers and are model agnostic: http://proceedings.mlr.press/v89/schulam19a/schulam19a.pdf … @_beenkim also has a paper along these lines: https://arxiv.org/abs/1805.11783
Important challenge for #DeepLearning is uncertainty quantification: knowing what it doesn't know. Neural rendering model adds latent variables to convolutional #NeuralNetworks and is accurate for out-of-distribution detection. @YujiaHuangC @dai_sh29 @TanNguyen689 Rich Baranuikhttps://twitter.com/arXiv_Daily/status/1149319305793638403 …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
CC
@david_sontag Quantifying uncertainty is critical for any decision support tool designed for clinicians. I would not trust a model’s recommendations if it lacked uncertainty parametersThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
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.