Timeline for Choosing $k$ different assignments of binary variables in order to capture the largest volume of the joint probability distribution
Current License: CC BY-SA 4.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Sep 23, 2022 at 7:58 | vote | accept | JanHula | ||
| S Sep 23, 2022 at 7:58 | vote | accept | JanHula | ||
| S Sep 23, 2022 at 7:58 | |||||
| S Sep 18, 2022 at 21:18 | vote | accept | JanHula | ||
| S Sep 23, 2022 at 7:58 | |||||
| S Sep 18, 2022 at 20:14 | vote | accept | JanHula | ||
| S Sep 18, 2022 at 21:18 | |||||
| S Sep 18, 2022 at 20:14 | vote | accept | JanHula | ||
| S Sep 18, 2022 at 20:14 | |||||
| S Sep 18, 2022 at 17:39 | vote | accept | JanHula | ||
| S Sep 18, 2022 at 20:14 | |||||
| Sep 18, 2022 at 17:38 | vote | accept | JanHula | ||
| S Sep 18, 2022 at 17:39 | |||||
| Sep 17, 2022 at 15:42 | answer | added | RobPratt | timeline score: 0 | |
| Sep 17, 2022 at 13:23 | answer | added | kodlu | timeline score: 0 | |
| Sep 17, 2022 at 6:15 | comment | added | JanHula | Another way to imagine this problem is to imagine an $n$-dimensional unit cube which is cut to pieces by doing a cut along each of the $n$ sides of the cube according to the proportion $p_i$. Then our task would be to choose $k$ different pieces to maximize the volume of the chosen pieces. | |
| Sep 17, 2022 at 6:14 | comment | added | JanHula | I assume that the joint PMF is a product of the marginals $\Pi p_i(x_i)$. I do not assume anything about the marginals but we could assume some restrictions if it would help. If $k = 1$, then the best choice is to choose the assignment with the highest joint probability. Which assigns the most probable value to each variable (according to $p_i$). | |
| Sep 17, 2022 at 5:24 | history | edited | RobPratt | CC BY-SA 4.0 |
deleted 1 character in body
|
| Sep 16, 2022 at 20:40 | comment | added | RobPratt | Do you know anything about the joint PMF a priori? Also, consider even the simplest case $k=1$. How do you measure how much a single assignment "captures" the distribution? | |
| Sep 16, 2022 at 20:32 | history | edited | RobPratt | CC BY-SA 4.0 |
added 2 characters in body
|
| Sep 16, 2022 at 20:10 | history | edited | YCor | CC BY-SA 4.0 |
formatting
|
| S Sep 16, 2022 at 20:01 | review | First questions | |||
| Sep 17, 2022 at 8:42 | |||||
| S Sep 16, 2022 at 20:01 | history | asked | JanHula | CC BY-SA 4.0 |