Timeline for can we trust feature importance of a poor model
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 21, 2021 at 5:33 | comment | added | Tim | @haneulkim you have results of an experiment in the linked thread, for importance in general, it is calculated using some model, so it cannot be independent of the underlying model. | |
| Oct 21, 2021 at 2:40 | comment | added | haneulkim | Thanks! Is there reference that proves your statement? | |
| Oct 21, 2021 at 2:39 | vote | accept | haneulkim | ||
| Oct 20, 2021 at 12:27 | history | edited | Tim | CC BY-SA 4.0 |
edited body
|
| Oct 20, 2021 at 9:32 | history | answered | Tim | CC BY-SA 4.0 |