Timeline for Is it possible for an ML model to 'overfit' training data but perform fairly well on val set, i.e generalize well too
Current License: CC BY-SA 4.0
5 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Mar 6, 2023 at 6:14 | vote | accept | Omer | ||
Aug 6, 2021 at 9:16 | comment | added | Tim | Johnny Lomond gave a nice answer (+1), but also for a trivial counterexample: say that by chance your validation set is very similar to the training set; even if the model badly overfits, by chance it will perform nicely on this particular validation set. | |
Aug 6, 2021 at 9:05 | answer | added | Jonny Lomond | timeline score: 2 | |
Aug 6, 2021 at 7:48 | review | First posts | |||
Aug 6, 2021 at 13:49 | |||||
Aug 6, 2021 at 7:47 | history | asked | Omer | CC BY-SA 4.0 |