Timeline for Given enough time, can a fully connected layer approximate a causal convolution layer?
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 31, 2023 at 15:47 | comment | added | Firebug | The Feed Forward network is a universal approximator, so it is possible to have one that can perfectly mimic the Wavenet in the local neighborhood of the data, given assumptions. That doesn't mean, however, that it's possible to learn this approximation by gradient descent. | |
| Mar 31, 2023 at 15:27 | history | edited | Dave |
edited tags
|
|
| Mar 31, 2023 at 15:20 | answer | added | SQLServerSteve | timeline score: 2 | |
| Mar 31, 2023 at 12:18 | history | edited | Dave |
edited tags
|
|
| Feb 20, 2023 at 17:50 | vote | accept | AlanSTACK | ||
| Feb 13, 2023 at 4:17 | history | edited | Sycorax♦ | CC BY-SA 4.0 |
edited title
|
| Feb 13, 2023 at 4:17 | answer | added | Dave | timeline score: 1 | |
| Feb 13, 2023 at 4:13 | history | asked | AlanSTACK | CC BY-SA 4.0 |