Timeline for Interpretation of covariance term in loss function
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 17, 2020 at 16:46 | comment | added | Prototank | Thank you for your thorough answer as well as the references. You expressed things very cleanly! | |
| Jun 17, 2020 at 16:45 | vote | accept | Prototank | ||
| Jun 16, 2020 at 21:25 | comment | added | chang_trenton | Yes, indeed, the multivariate normal PDF replaces the $\frac{x - \mu}{\sigma}$ in the exponential with $(x-\mu)^\top \Sigma^{-1} (x - \mu)$; hence, $p(x)$ in a multivariate normal is a function of the Mahalanobis distance. | |
| Jun 16, 2020 at 21:21 | comment | added | Firebug | This term also appears in the Multivariate Normal log-likelihood, so there's another interpretation possible from there. | |
| Jun 16, 2020 at 21:08 | history | edited | chang_trenton | CC BY-SA 4.0 |
clarified preliminaries
|
| Jun 16, 2020 at 21:03 | history | answered | chang_trenton | CC BY-SA 4.0 |