Timeline for contaminate data with label then take it away
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 2, 2020 at 4:58 | comment | added | Cryo | @Tim. Makes sense. Thanks | |
| Sep 1, 2020 at 5:18 | comment | added | Tim | but if model had access to the labels at training time, it can simply copy them while ignoring other variables. It does not need to learn anything. So it gives you no guarantees that what it learned is useful. | |
| Aug 31, 2020 at 20:25 | history | edited | Cryo | CC BY-SA 4.0 |
added 774 characters in body
|
| Aug 31, 2020 at 20:17 | comment | added | Cryo | @Tim I thought it could be useful as a sort of initialization strategy. What I meant was that in successive trainings you do not re-initialize the weights of the model, you start with previous ones. | |
| Aug 31, 2020 at 18:44 | comment | added | Tim | How would that be ever useful? You cannot use labels on prediction stage.You want to predict something that you don't know, otherwise why to predict it? If that is the case, then the "feature" would not be useful for you. | |
| Aug 31, 2020 at 18:03 | answer | added | Davis70 | timeline score: 3 | |
| Aug 31, 2020 at 13:46 | history | asked | Cryo | CC BY-SA 4.0 |