# Vector Embeddings - [Word embedding | Wikipedia](https://en.wikipedia.org/wiki/Word_embedding) - Any text can have embeddings, not just words. e.g. snippets in [[rag|RAG]]. ## Intro - Recommendation Systems, NLP, Computer Vision, Gen AI, LLM, etc are all based on vector embeddings. - Items put into an embedding space. - Making recommendations based on distance between vectors, e.g. _cosine distance_ $ \frac{A\cdot B}{\Vert A\Vert \Vert B\Vert} = \frac{\displaystyle\sum_{i=1}^n A_iB_i} { \sqrt{\displaystyle\sum_i^n A_i^2} \sqrt{\displaystyle\sum_i^n B_i^2} } $ - Examples of embedding dimensions: male-female, verb tense, country-capital - _Latent Space_, aka _Embedding Space_ - Word Embedding Models - Word2Vec - GloVe - BERT - GPT - VGGNet (image) - GoogLeNet (image) - [[nn|Neural Networks]] to obtain embedding