Timeline for what does the Wasserstein distance between two distributions quantify
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 4, 2020 at 21:23 | comment | added | Tobsn | Think of the real line. Are 1 and 5 close? | |
| Jun 28, 2020 at 23:22 | vote | accept | Steve | ||
| Jun 25, 2020 at 13:58 | answer | added | harwiltz | timeline score: 5 | |
| Jun 25, 2020 at 12:08 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| May 24, 2020 at 2:45 | answer | added | Charly Empereur-mot | timeline score: 5 | |
| Mar 24, 2020 at 16:08 | comment | added | Steve | Thank you for your answer, I understand the conceptual meaning of what this metric quantifies (as I mentioned in the post) but I would like to understand what the numbers really mean and how would I know that my distributions are close ? (for instance does 1 mean that the the pds close ? or 0.1 ? or maybe 0.001 ? and does the size of the two samples matter in this case ? for instance if we have two samples coming from two pds and each one of them has size 10000 and the distance between them is 1, does that mean they are close ? versus two samples with size 100 and have the same distance? ) | |
| Mar 24, 2020 at 15:59 | comment | added | jbowman | The Wikipedia page en.wikipedia.org/wiki/Wasserstein_metric is helpful in this regard; basically, the metric quantifies how much mass must be moved around, and how far, to turn one distribution into the other. | |
| Mar 24, 2020 at 14:37 | history | asked | Steve | CC BY-SA 4.0 |