Top all-time · #1896
Edwin Chen
Personal blog · blog.echen.medataai
Edwin Chen's personal blog at blog.echen.me covers data and ai. Since 2011, 41 different Hacker News readers have linked it 51 times, most often “Exploring LSTMs”. No feed was found, so read it on the site itself.
Most-linked posts?
- Exploring LSTMs6 people · 7 mentions · first linked 2017-05-31
- Introduction to Latent Dirichlet Allocation(2011)6 people · 6 mentions · first linked 2011-10-31
- Winning the Netflix Prize: A Summary5 people · 5 mentions · first linked 2011-10-26
- Stuff Harvard, MIT, Stanford & Caltech People Like3 people · 3 mentions · first linked 2011-11-09
- Infinite Mixture Models with Nonparametric Bayes and the Dirichlet Process3 people · 3 mentions · first linked 2012-03-21
- Introduction to Conditional Random Fields3 people · 3 mentions · first linked 2016-03-29
- Movie recommendations and more via MapReduce and Scalding2 people · 2 mentions · first linked 2012-02-29
- Tips and best practices for Mechanical Turk2 people · 2 mentions · first linked 2012-04-26
- A Quick Introduction to ggplot22 people · 2 mentions · first linked 2012-09-07
- Choosing a Machine Learning Classifier2 people · 2 mentions · first linked 2012-09-18
What Hacker News says?
For those interested in why and how, there's also http://blog.echen.me/2011/10/24/winning-the-netflix-prize-a-...
Comment linking /2011/10/24/winning-the-netflix-prize-a-summary/
on “Nate Silver correctly called 50 out of 50 states”Submitted Exploring LSTMs as a story · 349 points
HN discussionMy personal site is https://jonathanstreet.com/ It's in need of some help so I've been looking around for inspiration. I'll be taking a look at everything posted. Some I've found previously I like: http://blog.echen.me/ https://www.liyaoxiang.com/ http://www.randalolson.com/ https://jvns.ca/ http://www.nowozin.net/sebastian/ http://www.emilio.ferrara.name/ http://danielnee.com/ http://daynebatten.com/ https://thomas…
Comment linking /
on “Building a Kickass Portfolio”Submitted Hacker News Analysis as a story · 236 points
HN discussionWe know these things about Twitter relating to censorship: They have shut down Twitter accounts for abusing the TOS, and this process is on a case-by-case basis, and can affect well-known users such as Milo Yiannopoullous and Martin Shkreli: http://www.breitbart.com/milo/2016/07/19/breaking-milo-suspe... http://www.breitbart.com/tech/2017/01/08/pharma-legend-marti... Twitter also deletes content in response to valid…
Comment linking /2013/01/08/improving-twitter-search-with-real-time-human-computation
on “Twitter allegedly deleting negative tweets about United Airlines incident”Submitted Stuff Harvard, MIT, Stanford & Caltech People Like as a story · 94 points
HN discussionBack in August I had a look at using Latent Dirichlet Allocation ( http://blog.echen.me/2011/08/22/introduction-to-latent-diric... ) for improving how I auto-tag articles. It looked promising but I haven't deployed it yet. There are various recommendation approaches I'd like to try. If the site had more users (beyond the ~4 friends I showed it to) I'd like to try out a collaborative filtering approach ( https://en.w…
Comment linking /2011/08/22/introduction-to-latent-dirichlet-allocation/
on “Show HN: Blogosphere – Discover independent technical blogs”Submitted Stuff Harvard people Like as a story · 87 points
HN discussion- nl · 2016-03-30
Many, many startups use Amazon Mechanical Turk and/or CrowdFlower for this exact thing. See http://blog.echen.me/2012/04/25/making-the-most-of-mechanica... for some examples.
Comment linking /2012/04/25/making-the-most-of-mechanical-turk-tips-and-best-practices/
on “How to create an AI startup – convince some humans to be your training set” Submitted Creating a friend recommendation algorithm from social network data as a story · 59 points
HN discussionPrediction accuracy cannot be computed in a purely unsupervised setting: you don't have labels for the samples. You can compare the cluster found by k-means run several times against what you get with a randomized version of your dataset though: http://blog.echen.me/2011/03/19/counting-clusters/
Comment linking /2011/03/19/counting-clusters/
on “Machine Learning Cheat Sheet for scikit-learn”Submitted Introduction to Latent Dirichlet Allocation(2011) as a story · 8 points
HN discussionIt's a great text, but I recommend supplementing it with: http://blog.echen.me/2017/05/30/exploring-lstms/ . Additionally, GRU may be simpler to understand memory unit.
Comment linking /2017/05/30/exploring-lstms/
on “Understanding LSTM Networks (2015)”Submitted Movie recommendations and more via MapReduce and Scalding as a story · 6 points
HN discussionFor people interested in the algorithms that won the Netflix Prize (including the Matrix Factorization and Restricted Boltzmann Machine approaches mentioned in the first section), there's a summary here: http://blog.echen.me/2011/10/24/winning-the-netflix-prize-a-...
Comment linking /2011/10/24/winning-the-netflix-prize-a-summary/
on “Netflix Recommendations: Beyond the 5 stars”Submitted Choosing a Machine Learning Classifier as a story · 4 points
HN discussionHumans in the loop is great for this (kinda why mechanical turk was invented in the first place). I like echen's blog here http://blog.echen.me/2014/10/07/moving-beyond-ctr-better-rec...
Comment linking /2014/10/07/moving-beyond-ctr-better-recommendations-through-human-evaluation/
on “Item2Vec: Neural Item Embedding for Collaborative Filtering”Submitted Infinite Mixture Models with Nonparametric Bayes and the Dirichlet Process as a story · 4 points
HN discussionOooooh man, Latent Dirichlet Allocation is cool stuff, especially in the context of topic modeling (which is what Chen is doing here). OP actually wrote a pretty accessible blog post about how he does this sort of thing which you can see at [1]. If you don't want to read it, Mike Jordan has a pretty neat presentation about it at [2]. If you're statistically trained and don't want to view the video, you will probably…
Comment linking /2011/08/22/introduction-to-latent-dirichlet-allocation/
on “Stuff Harvard people Like”Submitted Tweets vs. Likes: What gets shared on Twitter vs. Facebook? as a story · 4 points
HN discussionProvenance of data was not in scope, 'twas more of a standard datamining "see if you can dig up something interesting" project. Like I said, there were scores of these - one of my colleagues wrote the famous soda vs pop thingy which once again put location stats to good use- http://blog.echen.me/2012/07/06/soda-vs-pop-with-twitter/
Comment linking /2012/07/06/soda-vs-pop-with-twitter/
on “Twitter Takes Aim at Anonymous Egg Accounts”Submitted Winning the Netflix Prize: A Summary as a story · 4 points
HN discussionIf it can be interfered from words alone, try Latent Dirichlet Allocation (e.g. with http://radimrehurek.com/gensim/ ) to generate tags. Some sources: * http://blog.echen.me/2011/08/22/introduction-to-latent-diric... * http://alexperrier.github.io/jekyll/update/2015/09/04/topic-... * http://engineering.flipboard.com/2017/02/storyclustering Alternatively, if you know tags, just want to see which are similar to each o…
Comment linking /2011/08/22/introduction-to-latent-dirichlet-allocation/
on “Ask HN: Any tools to easily cluster ideas/notes by similarity?”Submitted Exploring LSTMs as a story · 3 points
HN discussionWhy does anything called "impatient" start with anything else but direct code examples? (Cf. http://blog.echen.me/2012/01/17/quick-introduction-to-ggplot... or https://www.meteor.com/tutorials ) I am impatient and I prefer starting form direct minimalistic code samples (rather than, say, variable types). When I child starts learning LEGO blocks, its rarely from the abstract rules - but from minimalistic example. And…
Comment linking /2012/01/17/quick-introduction-to-ggplot2/
on “Impatient R”Submitted Winning the Netflix Prize: A Summary as a story · 3 points
HN discussionIf you're interested in scalding, Edwin Chen's awesome recommendation post can't be beat: http://blog.echen.me/2012/02/09/movie-recommendations-and-mo...
Comment linking /2012/02/09/movie-recommendations-and-more-via-mapreduce-and-scalding/
on “Why Hadoop MapReduce needs Scala”Submitted Stuff Harvard People Like as a story · 3 points
HN discussionI really enjoyed playing with BertViz. Another similar visual exploration tool I found recently is in Edwin Chen's blog. I am pretty sure this is the best explanation of LSTM. I think more tutorials should use this visual approach. http://blog.echen.me/2017/05/30/exploring-lstms/ There is also https://playground.tensorflow.org/
Comment linking /2017/05/30/exploring-lstms/
on “Show HN: Fully client-side GPT2 prediction visualizer”Submitted Stuff Harvard People Like as a story · 3 points
HN discussion
