Timeline for How to stop a neural network from regressing to the mean
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 12, 2020 at 10:40 | vote | accept | Alexander Soare | ||
| Sep 10, 2020 at 16:28 | history | edited | Alexander Soare | CC BY-SA 4.0 |
added 327 characters in body
|
| Sep 10, 2020 at 9:50 | comment | added | Alexander Soare | @Dave good point. I only have 45k params, but only 170 training samples (stupidly small I know - they are 3D CAT scans of lungs). I would have expected that for the training set at least I'd get the NN to look at a small handful of pixels and perfectly overfit | |
| Sep 10, 2020 at 3:05 | comment | added | Dave | @Graph4Me I’d still expect thousands, maybe millions, or parameters in a CNN to play connect-the-dots and overfit like crazy, even if the predictors and outputs are unrelated. | |
| Sep 9, 2020 at 22:24 | answer | added | Michał Słodki | timeline score: 1 | |
| Sep 9, 2020 at 20:03 | comment | added | Graph4Me Consultant | May be there is no causility as expected. So if its not possible to conclude the output, given the input, the best a model can do is to output the mean. | |
| Sep 9, 2020 at 19:55 | history | asked | Alexander Soare | CC BY-SA 4.0 |