177: Vector Databases

Описание: Intro topic:  Buying a Car




News/Links:



• Cognitive Load is what Matters
• https://github.com/zakirullin/cogniti...
• Diffusion models are Real-Time Game Engines
• https://gamengen.github.io/
• Your Company Needs Junior Devs
• https://softwaredoug.com/blog/2024/09...
• Seamless Streaming / Fish Speech / LLaMA Omni
• Seamless: https://huggingface.co/facebook/seaml...
• Fish: https://github.com/fishaudio/fish-spe...
• LLaMA Omni: https://github.com/ictnlp/LLaMA-Omni 




Book of the Show

• Patrick: 
• Thought Emporium Youtube
•    • Tactical Thermite Powered Hot Dog  
• Jason: 
• Novel Minds
• https://www.novelminds.ai/







Patreon Plug https://www.patreon.com/programmingth...






Tool of the Show

• Patrick: 
• Escape Simulator
• https://pinestudio.com/games/escape-s...
• Jason: 
• Cursor IDE
• https://www.cursor.com/



Topic: Vector Databases (~54 min)



• How computers represent data traditionally
• ASCII values
• RGB values
• How traditional compression works
• Huffman encoding (tree structure)
• Lossy example: Fourier Transform & store coefficients
• How embeddings are computed
• Pairwise (contrastive) methods
• Forward models (self-supervised)
• Similarity metrics
• Approximate Nearest Neighbors (ANN)
• Sub-Linear ANN
• Clustering
• Space Partitioning (e.g. K-D Trees)
• What a vector database does
• Perform nearest-neighbors with many different similarity metrics
• Store the vectors and the data structures to support sub-linear ANN
• Handle updates, deletes, rebalancing/reclustering, backups/restores
• Examples
• pgvector: a vector-database plugin for postgres
• Weaviate, Pinecone 
• Milvus




★ Support this podcast on Patreon ★ (  / programmingthrowdown  )