4 ms·
Vector embedding is not an invention of the last decade. Featurization in ML goes back to the 60s - even deep learning-based featurization is decades old at a m
by _jayhack_ 1y ago
Vector embedding is not an invention of the last decade. Featurization in ML goes back to the 60s - even deep learning-based featurization is decades old at a minimum. Like everything else in ML this became much more useful with data and compute scale
- senderista 1y agoYup, when I was at MSFT 20 years ago they were already productizing vector embedding of documents and queries (LSI).
- jongjong 1y agoInteresting. Makes one think.
- senderista 1y agoTo be clear, LSA[1] is simply applied linear algebra, not ML. I'm sure learned embeddings outperform the simple SVD[2] used in LSA. [1] https://en.wikipedia.org/wiki/Latent_semantic_analysis https://en.wikipedia.org/wiki/Latent_semantic_analysis [2] https://en.wikipedia.org/wiki/Singular_value_decomposition https://en.wikipedia.org/wiki/Singular_value_decomposition