Timeline for Are gaussians with different moments far in total variation distance?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 28, 2019 at 15:08 | comment | added | Sandeep Silwal | Can you explain why the first equality holds? (why is the L1 norm equal to sup over functions that satisfy |f| <=1 ?) | |
| Mar 24, 2014 at 14:38 | history | bounty ended | Paul Christiano | ||
| Mar 19, 2014 at 10:49 | comment | added | George Lowther | I switched the norm used for $u$ to fix the minor issue | |
| Mar 19, 2014 at 10:44 | history | edited | George Lowther | CC BY-SA 3.0 |
edited body
|
| Mar 19, 2014 at 3:04 | vote | accept | Paul Christiano | ||
| Mar 19, 2014 at 3:04 | comment | added | Paul Christiano | Thanks! I'll give this the bounty unless a significantly cleaner solution crops up (unless there is some etiquette about this I don't know). Minor (I think) issue: the covariance matrices can have eigenvalues more than 1. But you only need to apply this bound for sparse vectors, so it looks like you are good. | |
| Mar 18, 2014 at 23:54 | history | edited | George Lowther | CC BY-SA 3.0 |
added 6 characters in body
|
| Mar 18, 2014 at 23:37 | history | edited | George Lowther | CC BY-SA 3.0 |
deleted 9 characters in body
|
| Mar 18, 2014 at 23:29 | history | answered | George Lowther | CC BY-SA 3.0 |