Timeline for Relation between information geometry and geometric deep learning
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 13, 2022 at 16:13 | vote | accept | Blupon | ||
| Dec 20, 2020 at 17:07 | comment | added | Martin M. W. | That's a good point! (And more generally, one might add that information geometry has a life beyond neural networks--this summary is just focusing on its intersections with deep learning.) | |
| Dec 20, 2020 at 15:40 | comment | added | Gabe K | This is a good overview. It should be noted that it's possible to do information geometry on non-Euclidean spaces, it's just that most of the common examples and applications (e.g., exponential families) are defined on domains in Euclidean space. | |
| Dec 19, 2020 at 19:40 | history | edited | Martin M. W. | CC BY-SA 4.0 |
Clarified a few points, realized "Euclidean" was confusing in one sentence.
|
| Dec 19, 2020 at 14:23 | history | answered | Martin M. W. | CC BY-SA 4.0 |