Morning thought: 2x2 diagrams work so well because a random pair of vectors in a high dimensional space are orthogonal with probability ~1 @vgr
-
-
Replying to @eigenstate @vgr
I don't understand this. Please explain me like I'm a child or share some links.
1 reply 0 retweets 6 likes -
Same. New twitter idea, post these thoughts and then ELI5.
1 reply 0 retweets 0 likes -
I had no idea what ELI5 meant until now! I'll do a thread once I have an extended moment :)
1 reply 0 retweets 3 likes -
Looking forward to it!
1 reply 0 retweets 0 likes -
Abhishek Agarwal 📝 🖱️ 📸 Retweeted Blake Riley
The blog linked here explained in a very clear, simple manner.https://twitter.com/blakejriley/status/1293931550384553986?s=19 …
Abhishek Agarwal 📝 🖱️ 📸 added,
Blake Riley @blakejrileyReplying to @vgr @eigenstateWell-known result. Two ways to get there: 1) if the elements are iid and centered, apply LLN to have the inner product converge to zero 2) most of the area of a hypersphere is near the equator https://www.johndcook.com/blog/2017/07/13/concentration_of_measure/ …1 reply 0 retweets 0 likes -
Cook does an experiment. Just look at vgr's plots -- his explain just as much! There's a lot more mathematics there :D (I have a resentment towards cook's blog for doing ML on homotopy groups of spheres once because while a cool idea it just so completely misses the point...)
1 reply 0 retweets 0 likes
Oh, interesting! His graphs end up looking very like mine (I didn't see it before I did mine)
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.