Does anyone know how to find the optimal dimensionality for embedding-type representations? Mostly I'm hand-tuning
-
-
Replying to @BayesForDays
Would the same trick they used http://www-bcf.usc.edu/~gareth/research/ratedist.pdf … work for your case?
1 reply 0 retweets 1 like -
Replying to @sbelak
While a cool paper, I'm trying to figure out if I can capture high and low frequency word similarities with a single dimensionality
1 reply 0 retweets 0 likes
Replying to @BayesForDays
My thinking was if you can define something similar to their distortion for your problem, you might see the same dynamic
2:45 PM - 4 Sep 2017
0 replies
0 retweets
1 like
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.