Timeline for Why transformer in deep learning is called transformer?
Current License: CC BY-SA 4.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 13 at 17:46 | answer | added | root | timeline score: 1 | |
| Aug 13, 2023 at 19:17 | comment | added | Vlad | But those matrix multiplications are exactly linear. There is only softmax function applied on the top of scalar products of K and Q and then again linearly multiplied by V. | |
| Oct 15, 2022 at 14:42 | answer | added | Ариж Аль Адел | timeline score: 1 | |
| Sep 9, 2021 at 3:19 | comment | added | patagonicus | Probably because transformer architectures use series of (non)linear transformations, i.e., recall Q, V, K matrices. That might be the reason. @Tim yes fun name, if we associate with Transformers franchise. Interesting that name is not disputed by Hasbro. | |
| Aug 26, 2021 at 19:33 | review | Low quality posts | |||
| Aug 26, 2021 at 20:25 | |||||
| Aug 26, 2021 at 16:18 | comment | added | Arya McCarthy | Yeah, +1 to Tim. It’s marketing. | |
| Aug 26, 2021 at 16:13 | history | edited | kjetil b halvorsen♦ | CC BY-SA 4.0 |
edited title
|
| Aug 26, 2021 at 9:37 | comment | added | Tim | Someone thought it would be a fun name... | |
| Aug 26, 2021 at 9:18 | review | Low quality posts | |||
| Aug 26, 2021 at 13:25 | |||||
| Aug 26, 2021 at 9:04 | review | First posts | |||
| Aug 26, 2021 at 16:13 | |||||
| Aug 26, 2021 at 9:01 | history | asked | Johnny Tam | CC BY-SA 4.0 |