Connecting the dots in hyperspace: Why we built lance-csr
We introduce lance-csr, a fast graph query engine over Lance datasets that enables exact queries in Cypher or SQL, and associative search with hypervectors over a single copy of the data.
Topic
A gentle, visual introduction to hyperdimensional computing: encoding data, associative search, and relationships.
Showing 1–10 of 15 posts
We introduce lance-csr, a fast graph query engine over Lance datasets that enables exact queries in Cypher or SQL, and associative search with hypervectors over a single copy of the data.
When a robot's load, charge or temperature changes, a familiar route can make very different demands. We explore how HDC memory could help it prepare for the journey ahead.
A look back at how the HDC literature describes the terms "cleanup" and "crosstalk", and how we can view both terms through a modern lens using vector search in LanceDB.
An HDC experiment that separates stored examples from temporary class summaries, and compares rebuilding with caching.
Why shared bipolar binding preserves a ranking, how weighted similarity can change it, and what remains to be tested for useful retrieval.
How a compact abstraction can use thousands of coordinates, and what the extra width buys an HDC system.
Learn how to encode node properties and edge direction with HDC, compose a partial query, and resolve similarity matches back to exact graph records.
How HRR, BSC, and MAP implement the same broad VSA computation shape despite using different hyperspaces and mathematical operations.
Graph search follows exact relationships, while associative search finds similar situations from incomplete examples.
Why the future of AI may be hybrid: deep learning learns powerful representations, while HDC complements it with compositional operations, associative memory, and lightweight adaptation.
Contact
If you're working with connected data, retrieval, agent memory or online learning, we'd love to hear what you're building and where the current representation is falling short.
We usually reply within two business days.