Timeline for Why is binary cross entropy (or log loss) used in autoencoders for non-binary data
Current License: CC BY-SA 4.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 24, 2021 at 1:27 | answer | added | Kevin | timeline score: 1 | |
| Jul 30, 2020 at 16:13 | comment | added | mrgloom | There is a typo in calculations, but yes loss is asymmetric: >>> -0.2 * np.log(0.1) - (1.0-0.2) * np.log(1.0-0.1) 0.5448054311250702 >>> -0.2 * np.log(0.3) - (1.0-0.2) * np.log(1.0-0.3) 0.5261345160161732 | |
| Jun 10, 2020 at 14:52 | history | edited | Flek | CC BY-SA 4.0 |
Fixed a typo
|
| S Mar 8, 2019 at 14:29 | history | bounty ended | Flek | ||
| S Mar 8, 2019 at 14:29 | history | notice removed | Flek | ||
| Mar 8, 2019 at 14:27 | vote | accept | Flek | ||
| Mar 7, 2019 at 22:36 | answer | added | hans | timeline score: 8 | |
| Mar 7, 2019 at 22:02 | comment | added | hans | Sure, I was just joking :-) I understand your question. For binary data answer is obvious - gradient behaves as constant when prediction is very close to true value (for log), while for mse it is linear, i.e. it becomes small. | |
| Mar 7, 2019 at 19:00 | comment | added | Flek | Indeed, I used log10, but the base doesn't matter. What matters is the asymmetry and what its benefits are. Maybe log loss and mse are both equally useful. I was just wondering as it appeared to me that ~50% use log loss for image autoencoders and the other ~50% use mse. | |
| Mar 7, 2019 at 17:49 | comment | added | hans | Funny, I pasted your calculations into python and I got 0.639 and 0.526 respectively. Probably you used log based on 10 and I used natural logarithm. | |
| Mar 6, 2019 at 12:00 | history | tweeted | twitter.com/StackStats/status/1103264123419328512 | ||
| Mar 2, 2019 at 5:34 | answer | added | ReneBt | timeline score: 1 | |
| Mar 1, 2019 at 22:48 | answer | added | JaeHyeok Shin | timeline score: 3 | |
| S Mar 1, 2019 at 15:34 | history | bounty started | Flek | ||
| S Mar 1, 2019 at 15:34 | history | notice added | Flek | Draw attention | |
| Feb 28, 2019 at 14:43 | history | edited | Flek | CC BY-SA 4.0 |
Fixed a small typo to clarify the sentence
|
| Feb 26, 2019 at 23:55 | review | First posts | |||
| Feb 27, 2019 at 7:13 | |||||
| Feb 26, 2019 at 23:51 | history | asked | Flek | CC BY-SA 4.0 |