29 ms·

Reactive Machine Learning via Jeff Smith

• jeffreysmith 12y ago
  Jeff, here. I put this up in time for folks coming to Spark Summit East in the hope that the influx of Spark folks to NYC might want to think a bit about my favorite type of distributed system: machine learning. Even though this thing is already pretty lengthy, I still feel like there's a lot more that could be explored here. See the open questions at the end of the post. Would love to hear the feedback from the HN folks about designing reactive machine learning systems.
• newman8r 12y ago
  The example they use is hilarious and makes me think about one of my ideas for a dog crap cleaning backyard quadruped. It's not quite as good as that spray in http://www.imdb.com/title/tt0326856/ http://www.imdb.com/title/tt0326856/ though lol. terrible movie.
  • jeffreysmith 12y ago
    I do like to use dogs in my examples. See https://medium.com/tractable-constraints/in-search-of-diversity-41f791e0ddf0 https://medium.com/tractable-constraints/in-search-of-divers... and https://medium.com/j-smith-in-ink/the-purginator-1dfae16327b0 https://medium.com/j-smith-in-ink/the-purginator-1dfae16327b... . There's something about canines and k-means that just makes sense to me.
    • amsheehan 12y ago
      I wish I could get my k-nearest neighbors dog to shut up already.