Timeline for Are the raw probabilities obtained from XGBoost, representative of the true underlying probabilties?
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 1, 2020 at 23:56 | comment | added | dant | Old qustion, but for posterity: modern discriminative neural networks are definitively NOT calibrated in the sense that the OP posed (if they were, a big motivation for investigating generative networks would be redundant). See these papers: arxiv.org/pdf/1706.04599.pdf, papers.nips.cc/paper/… | |
| S Aug 7, 2019 at 7:19 | history | suggested | JoeyC | CC BY-SA 4.0 |
fixed minor typos
|
| Aug 7, 2019 at 5:41 | review | Suggested edits | |||
| S Aug 7, 2019 at 7:19 | |||||
| Oct 12, 2018 at 16:45 | vote | accept | Gale | ||
| Mar 8, 2018 at 14:31 | vote | accept | Gale | ||
| Oct 12, 2018 at 16:45 | |||||
| Mar 8, 2018 at 14:28 | history | edited | aivanov | CC BY-SA 3.0 |
added 228 characters in body
|
| Mar 8, 2018 at 14:07 | comment | added | aivanov | I've edited the answer: one would need another reliability plot (based on calibrated output). Note that you cannot do this assessment for every single person, only for the group of people who land in one bin (in terms of prediction). Also, according to my understanding, large number of bins could lead to poor results. | |
| Mar 8, 2018 at 14:06 | history | edited | aivanov | CC BY-SA 3.0 |
added 250 characters in body
|
| Mar 8, 2018 at 13:51 | comment | added | Gale | After calibration, how would one be able to confirm that the probabilities obtained, are approximately representative ? Since predicting the likelihood that someone would develop a disease, 55% chance as compared with a 70% is a significant difference, and thus comes the importance of accurately predicting these probabilities. | |
| Mar 8, 2018 at 13:24 | history | answered | aivanov | CC BY-SA 3.0 |