The Super Effectiveness of Pokémon Embeddings Using Only Raw JSON and Images

Re:The Super Effectiveness of Pokémon Embeddings Using Only ... by PaulHoule on Sunday June 30, 2024 @04:06PM (#40839917)

Also you hear that example over and over again because you can't get other ones to work reliably with Word2Vec; you'd have thought you could train a good classifier for color words or nouns or something like that if it worked but actually you can't.

Because it could not tell the difference between word senses I think Word2Vec introduced as many false positives as true positive, BERT was the revolution we needed.

I use similar embedding models for classification and it is great to see improvements in this space.

[ Parent | on HN ]

Re:The Super Effectiveness of Pokémon Embeddings Using Only ... by simonw on Sunday June 30, 2024 @06:16PM (#40840833)

The other example that worked for me with Word2Vec was Germany + Paris - France = Berlin: https://simonwillison.net/2023/Oct/23/embeddings/#exploring-...

[ Parent | on HN ]

Re:The Super Effectiveness of Pokémon Embeddings Using Only ... by craigacp on Sunday June 30, 2024 @09:20PM (#40841793)

There are a bunch of these things in a word2vec space. I had a blog post years ago on my group's blog which trained word2vec on a bunch of wikias so we could find out who is the Han Solo of Doctor Who (which I think somewhat inexplicably was Rory Williams). You need to carefully implement word2vec, and then the similarity search, but there are plenty of vaguely interesting things in there once you do.

[ Parent | on HN ]

Re:The Super Effectiveness of Pokémon Embeddings Using Only ... by radarsat1 on Monday July 01, 2024 @05:08AM (#40843928)

It's a good point about true and false positives though, which makes me wonder if anyone's taken a large database of expected outputs from such "equations" and used it to calculate validation scores for different models in terms of precision and recall.

[ Parent | on HN ]