| (5 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
| − | <center><font size= 4> | + | <center><font size="4"></font> |
| − | [[Maximum Likelihood Estimation | + | <font size="4">Questions and Comments for: [[Video slecture in English: Introduction to Maximum Likelihood Estimation|Introduction to Maximum Likelihood Estimation]] </font> |
| − | </font | + | |
| − | A [https://www.projectrhea.org/learning/slectures.php slecture] by Anantha Raghuraman | + | A [https://www.projectrhea.org/learning/slectures.php slecture] by Anantha Raghuraman |
| + | </center> | ||
| + | ---- | ||
| − | |||
| − | |||
---- | ---- | ||
| + | |||
| + | This is the discussion page for the slecture on [[Video slecture in English: Introduction to Maximum Likelihood Estimation|Introduction to Maximum Likelihood Estimation]]. Please leave me a comment below if you have any questions, or if you would like to discuss a topic. | ||
| + | |||
---- | ---- | ||
| + | = Review by ? = | ||
| − | + | [Review by Wen Yi] | |
| + | This slecture makes a fundamental introduction about Maximum Likelihood Estimation. | ||
| − | + | For the first half of the vedio, it focus on explaining the meaning of likelihood using Bernoulli distribution. It's a really clear expaination and quite worthwhile, as many people have difficulty in understanding the relationship between jiont probability of the samples' distribution and the likelihood. | |
| − | + | For the second half of the vedio, the author focus mainly on how to estimate the most popular distribution, Gaussian distribution. Inside this part, the author also mentioned some practical method like log-likelihood, derivation to make the estimation easier. At the end of the vedio, the author also talks about when would MLE work or not work. | |
| + | |||
| + | Good points: | ||
| + | |||
| + | I like the explanation of likelihood. It can be understand by my friend who is now from this area, that's good for a beginner on MLE. Just a small catch on the vedio, the example used should be drop a dice, not die.:) | ||
| + | |||
| + | The proof of the Gaussian MLE method is a very detail one, thus quite persuasive. | ||
| + | |||
| + | Some suggestions: | ||
| + | |||
| + | Although it's not so easy for a non-native speaker, I think people will prefer a vedio with a faster pace. While watching it, sometimes I just had to skip several second to let the author finish writing the whole equation or explain another example for the same kind of case. If improve the speed a bit, I think the whole slecture can be finished within 30 minutes, which can be cut into 2 part. | ||
| + | |||
| + | For the end of the vedio, the author mention using the wrong kind of distribution method for MLE will lead the performance dropped, it's easier to make people understand it with just a simple case (for example using Bernoulli on Gaussian destribution). | ||
---- | ---- | ||
| − | [[ | + | |
| + | [[2014 Spring ECE 662 Boutin|Back to ECE 662 S14 course wiki]] | ||
[[ECE662|Back to ECE 662 course page]] | [[ECE662|Back to ECE 662 course page]] | ||
Latest revision as of 08:12, 13 May 2014
Questions and Comments for: Introduction to Maximum Likelihood Estimation
A slecture by Anantha Raghuraman
This is the discussion page for the slecture on Introduction to Maximum Likelihood Estimation. Please leave me a comment below if you have any questions, or if you would like to discuss a topic.
Review by ?
[Review by Wen Yi]
This slecture makes a fundamental introduction about Maximum Likelihood Estimation.
For the first half of the vedio, it focus on explaining the meaning of likelihood using Bernoulli distribution. It's a really clear expaination and quite worthwhile, as many people have difficulty in understanding the relationship between jiont probability of the samples' distribution and the likelihood.
For the second half of the vedio, the author focus mainly on how to estimate the most popular distribution, Gaussian distribution. Inside this part, the author also mentioned some practical method like log-likelihood, derivation to make the estimation easier. At the end of the vedio, the author also talks about when would MLE work or not work.
Good points:
I like the explanation of likelihood. It can be understand by my friend who is now from this area, that's good for a beginner on MLE. Just a small catch on the vedio, the example used should be drop a dice, not die.:)
The proof of the Gaussian MLE method is a very detail one, thus quite persuasive.
Some suggestions:
Although it's not so easy for a non-native speaker, I think people will prefer a vedio with a faster pace. While watching it, sometimes I just had to skip several second to let the author finish writing the whole equation or explain another example for the same kind of case. If improve the speed a bit, I think the whole slecture can be finished within 30 minutes, which can be cut into 2 part.
For the end of the vedio, the author mention using the wrong kind of distribution method for MLE will lead the performance dropped, it's easier to make people understand it with just a simple case (for example using Bernoulli on Gaussian destribution).
