Timeline for Activation functions for autoencoder performing regression
Current License: CC BY-SA 3.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 26, 2018 at 10:41 | comment | added | shimao | It's possible that using linear instead of relu at the end helps the gradients flow better and avoids the dying relu problem, so I wouldn't be too surprised. | |
| Mar 26, 2018 at 10:38 | comment | added | RiccB | I tried several parameters and, almost independently of the scaler used (including no scaler), a linear activation in the output layer works almost always better than relu. Which I find odd. | |
| Mar 26, 2018 at 10:07 | history | answered | shimao | CC BY-SA 3.0 |