Hacker News Reader

Comment by mwcremer | Hacker News Reader

It feels to me like there is no way to "improve" short of blindly guessing synonyms. The space feels like concentric circles, where two words with similar word2vec scores move along the circumference rather than normal or even tangential.

View on HN

Try Pimantle[1], which is Semantle with a point-cloud visualization of guesses. The visualization is only two-dimensional, but it's enough to prevent the "orbiting" you describe.

[1]: https://semantle.pimanrul.es/

I do like this because you can see if the "path" you are on towards the center is going to get you close or if you are going to hit a dead end. Because it's weird some of the ways you can get relatively close to the word and then go nowhere.

I have also found it frustrating in semantle when I get something that is "close" but is actually close because it is a direct antonym of the word.

Only took me about 70 guesses including a dozen hints. I'm not good at thinking like these algorithms...

What dimensionality reduction algorithm is being used? PCA? UMAP? Of course better ones are likely to give better 2D representations.

I've had this same idea to use the point cloud visualization to improve my own take on language model games - but I when I experimented with this, I found that even the best 2D representations were still quite bad in the context of trying to help out humans.

PCA destroys local structure at such a low dimensionality. So probably UMAP or T-SNE.

My computer is on fire trying to run this. But it's way better than semantle

Ah, indeed! Not much data yet, but with my first pimantle just now, I was able to get it in 33 tries, instead of often over a hundred with semantle. It really helps to "see" how you are approaching (or not approaching) the word in one more dimension.

Reply on news.ycombinator.com