Timeline for Avoid overfitting in regression: alternatives to regularization
Current License: CC BY-SA 3.0
21 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 17, 2023 at 8:52 | answer | added | skan | timeline score: 0 | |
| Oct 15, 2020 at 9:26 | answer | added | Christabella Irwanto | timeline score: 2 | |
| Jul 20, 2017 at 19:34 | comment | added | Berk U. | To add to the comment by Digio: regularization is cheap compared to bagging/boosting but still expensive compared to the alternative of "no regularization" (see e.g. this post by Ben Recht on how regularization makes deep learning hard). If you have a huge number of samples, no regularization can work well for far cheaper. The model can still generalize well as @hxd1001 points out) | |
| Jul 20, 2017 at 19:25 | answer | added | Berk U. | timeline score: 3 | |
| Jul 20, 2017 at 18:10 | vote | accept | Benoit Sanchez | ||
| Jul 20, 2017 at 0:24 | history | rollback | gung - Reinstate Monica |
Rollback to Revision 3
|
|
| Jul 19, 2017 at 22:51 | comment | added | Dan | This might be of interest: stats.stackexchange.com/a/161592/40604 | |
| S Jul 19, 2017 at 22:37 | history | suggested | Ky - |
database, not machine learning, statistics, or any other major topic of this site
|
|
| Jul 19, 2017 at 21:34 | review | Suggested edits | |||
| S Jul 19, 2017 at 22:37 | |||||
| S Jul 19, 2017 at 20:35 | history | suggested | Mayou36 | CC BY-SA 3.0 |
small language changes
|
| Jul 19, 2017 at 20:08 | review | Suggested edits | |||
| S Jul 19, 2017 at 20:35 | |||||
| Jul 19, 2017 at 16:14 | history | tweeted | twitter.com/StackStats/status/887707129309667328 | ||
| Jul 19, 2017 at 14:20 | answer | added | Matthew Gunn | timeline score: 6 | |
| Jul 19, 2017 at 14:04 | comment | added | Digio | The advantage of regularization is that it's computationally cheap. Ensemble methods such as bagging and boosting (etc.) combined with cross validation methods for model diagnostics are a good alternative, but it will be a much more costly solution. | |
| Jul 19, 2017 at 13:32 | answer | added | Haitao Du | timeline score: 12 | |
| Jul 19, 2017 at 13:02 | answer | added | Ben Ogorek | timeline score: 15 | |
| Jul 19, 2017 at 12:56 | answer | added | Andrey Lukyanenko | timeline score: 2 | |
| Jul 19, 2017 at 12:05 | comment | added | seanv507 | Why not use norm regularisation? For neural networks , there is dropout | |
| Jul 19, 2017 at 10:47 | history | edited | Benoit Sanchez | CC BY-SA 3.0 |
added 67 characters in body
|
| Jul 19, 2017 at 10:45 | comment | added | Pere | "Big datasets" may mean a lot of observations, a lot of variables or both, and the answer may depend on the number of observations and variables. | |
| Jul 19, 2017 at 10:42 | history | asked | Benoit Sanchez | CC BY-SA 3.0 |