SOURCE-LINKED INTELLIGENCE

A Feature-Major Codebook for Memory-Efficient Sparse-Binary Self-Organizing Maps: Scaling a MEDLINE Atlas to 1.05 Million Neurons on a Single Consumer GPU

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Building a self-organising map at MEDLINE scale has been impractical: the best-matching-unit (BMU) search that dominates training is bound by the bandwidth needed to read the codebook every epoch. I show that this bottleneck is largely an artefact of codebook layout. Storing it feature-major with each feature's weights contiguous, W[v.M+i], recasts the search as a tiled sparse-dense product in which every loaded weight column is reused across a tile of samples. Varying only the layout, with implementation, precision and update rule held fixed, accelerates the BMU search by 4.5-8.5x, and becaus

Read original source ↗ Open in workspace

recordType

paper

region

Global