TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

A sane introduction to maximum likelihood estimation and maximum a posteriori

174 pointsby peroneover 6 years ago

5 comments

lenticularover 6 years ago
This is a really cool and clear introduction to MAP&#x2F;MLE, especially since you take great pains to explain what all of the notation means. I&#x27;ll definitely be pointing some people I know to this blog.<p>OT on technical blogs: Experts often are unable to put themselves in the shoes of someone with no experience, which really harms the pedagogy. When one practices a technical topic for a long time, concepts that were once foreign and difficult become instinctual. This makes it very hard to understand in what ways a beginner could be tripped up. It takes a large amount of thought to avoid this problem, which I think is why much introductory material - blog posts, books, etc., is really sub-par.
评论 #18814707 未加载
subjectHaroldover 6 years ago
Could someone explain in a bit more detail the move from 26 to 27? I don&#x27;t get the significance of being &quot;worried about optimization&quot; or why&#x2F;how we cancel p(x). I do get the later point about integration and the convenience of the reformulation. I just don&#x27;t get why or how it is &quot;allowed&quot;.<p>Sorry if this is obvious but I have been doing a lot of reading on this and have come across this step a few times before...but am just missing some part of every explanation.
评论 #18812639 未加载
评论 #18812615 未加载
kahoonover 6 years ago
Nice, clear explanation. Looking forward to the Bayesian inference one!<p>One note though: I think on equation 25 you are missing a log on the left hand side.
评论 #18811535 未加载
stilley2over 6 years ago
Nice write-up! Minor nitpick: ML&#x2F;MAP estimators don&#x27;t _require_ observations to be independent. At least, in my field we&#x27;re looking at a single observation of a multivariate distribution, and we don&#x27;t need to assume the elements are independent (ie, we permit a non-diagonal covariance matrix). My intuition says this is equivalent to assuming multiple correlated scaler observations, but I&#x27;d have to sit down with some paper. Also, you use &quot;trough&quot; where I think you mean &quot;through.&quot;
评论 #18812844 未加载
评论 #18812968 未加载
master_yoda_1over 6 years ago
So you think all the others are insane ;)
评论 #18812979 未加载