Clickbait title. Should be something like: efficiently find the most frequent items in a stream of data.
TL;DR: split your items by chunck of 8 bits. Store for each chunck A the number of time you saw it before the chunck B followed by C.
Then you keep track of only 2^24 counters instead of one counter by possible item.
Then using some graph algorithm (the interesting part, which is not explained in the article) you can use those weighted edges to reconstruct the most frequent items with a high confidence.
So this is essentially short complete Markov chain with recovery? What is the secret to performance in this probabilistic compressor? Word size alignment and locality?
Is this based on new extended variant of Ross's theorem? (multivariate case)
Thank you for the link. Interesting, they used the graph resampling and r-tree approaches already in use for certain kinds of machine learning. (Esp. approximate eigenvalue based classifiers)
Very well written too. Unfortunately the approach described only works for dense countable data that is linearly separable (eta spectral clusters, in fact even stronger assumption) - which still makes it quite useful as a database index.
If you know what's going on, the title might be clickbait. But I looked at the abstract (via the link in the article on Quanta Magazine) and understood none. Read the article and understood some. They explained the Big-O, the algorithm and the underlying graph that's being created. For me the bravour in the title is (just) fitting.
Upvoted your comment, but I worry this trend will only get stronger if we endorse and reinforce such behaviour. Why provide eyeballs, via the HackerNews front page?
The OP is truly disgusting clickbait and crossed the line for me.
7 comments
[ 2.8 ms ] story [ 22.2 ms ] threadTL;DR: split your items by chunck of 8 bits. Store for each chunck A the number of time you saw it before the chunck B followed by C. Then you keep track of only 2^24 counters instead of one counter by possible item. Then using some graph algorithm (the interesting part, which is not explained in the article) you can use those weighted edges to reconstruct the most frequent items with a high confidence.
Is this based on new extended variant of Ross's theorem? (multivariate case)
Very well written too. Unfortunately the approach described only works for dense countable data that is linearly separable (eta spectral clusters, in fact even stronger assumption) - which still makes it quite useful as a database index.
Also, the original paper is worth a read: https://arxiv.org/abs/1604.01357
The OP is truly disgusting clickbait and crossed the line for me.
I flagged it for this reason.