Timeline for Bayesian vs Maximum entropy
Current License: CC BY-SA 2.5
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 5, 2017 at 23:00 | history | edited | kjetil b halvorsen♦ |
edited tags
|
|
| Jan 20, 2012 at 18:15 | answer | added | Xi'an | timeline score: 7 | |
| Mar 9, 2011 at 16:10 | comment | added | robin girard | @Ashok most often connection you are searching for arises from the description of convex sets with a probability measure on its extreme points (Choquet Theory). | |
| Mar 9, 2011 at 14:19 | comment | added | Ashok | Robin, fact is that I don't know Bayesian inference method completely. Otherwise this question would not even have arisen to me. Now I am trying to find time to learn about Bayesian. All I knew(roughly) was that using Bayes theorem if some prior and some additional information are given one can update the probabilities. I do not know this rigorously. Whereas I know MaxEnt rigorously what it means. If it is possible, please explain or lead me(i.e.,point some reference) to learn Bayesian inference rigorously. Thank you. | |
| Mar 9, 2011 at 14:09 | history | edited | Ashok | CC BY-SA 2.5 |
deleted 176 characters in body
|
| Mar 9, 2011 at 12:22 | comment | added | robin girard | @Ashok Can you define formally what you mean by "the Bayesian rule of selection has a process of selecting the posterior given the prior which is supported by Bayes theorem." . | |
| Mar 3, 2011 at 11:33 | comment | added | Simon Byrne | You may find this question useful: stats.stackexchange.com/q/4978/495 | |
| Mar 3, 2011 at 11:33 | comment | added | Simon Byrne | My point was that to calculate D(P||Q), Q must be absolutely continuous with respect to P (so cannot be an arbitrary distributions). | |
| Mar 3, 2011 at 2:23 | comment | added | Ashok | May be, this is the difference between the two methods. I do not really know Bayesian method completely. As for MaxEnt $Q$ can be any distribution which is mostly a prior knowledge about the system (not the prior you are talking about). Pardon me if I am wrong. | |
| Mar 2, 2011 at 11:14 | comment | added | Simon Byrne | No, you said that Q was a prior. But a prior for what? If it is a prior for P, then it must be a distribution over E, in which case D(P||Q) doesn't really make sense. | |
| Mar 2, 2011 at 2:24 | comment | added | Ashok | Do you mean to ask, is $Q\in E$? Need not be. | |
| Mar 1, 2011 at 19:01 | comment | added | Simon Byrne | Is Q a distribution over E? | |
| Mar 1, 2011 at 18:55 | history | edited | Simon Byrne |
edited tags
|
|
| Mar 1, 2011 at 16:57 | history | edited | whuber♦ | CC BY-SA 2.5 |
added 7 characters in body
|
| Mar 1, 2011 at 14:23 | history | tweeted | twitter.com/#!/StackStats/status/42590693595693058 | ||
| Mar 1, 2011 at 12:01 | history | asked | Ashok | CC BY-SA 2.5 |