Timeline for Training a model purely on weak labels
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 3, 2021 at 2:40 | vote | accept | Blue482 | ||
| May 1, 2021 at 23:18 | answer | added | Erwan | timeline score: 2 | |
| May 1, 2021 at 15:15 | comment | added | Blue482 | I can't find the other paper. But here is one (arxiv.org/pdf/2101.09244.pdf). See Figure 1. I didn't post any particular paper because it seems to me a pattern across different studies, and the logic is the same and a clear one, i.e. training a ml classifier only on weak labels. | |
| May 1, 2021 at 14:11 | comment | added | Erwan | Please add links to the papers that you mention, otherwise it's hard to know exactly what this is about. Based only on your description I'd say that I agree with your interpretation, but it depends what the authors do exactly. | |
| May 1, 2021 at 2:15 | history | asked | Blue482 | CC BY-SA 4.0 |