Timeline for How 'deep' is deep for auto-encoders based deep learning?
Current License: CC BY-SA 3.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 17, 2016 at 8:09 | comment | added | hidemyname | arxiv.org/pdf/1512.03385v1.pdf Hi I saw the a thousand layer from this paper, although the a thousand layers is not better then merely a hundred layers from their experiments | |
| Sep 17, 2016 at 3:55 | comment | added | Danica | "while in image recognition, they use a thousand layers" – I definitely haven't seen any thousand-layer models. :) As far as I know (not exactly my area), they deepest common model in use in image recognition is ~30. | |
| Sep 16, 2016 at 10:45 | history | answered | hidemyname | CC BY-SA 3.0 |