Unsupervised learning · Morphogenetic Programming · UCL · 2025
A self-organising map, written from scratch in C#, that lays out the space between iconic buildings.
Method

01 Inputs

02 Normalise

03 Sample

04 Encode

05 Train & decode
Results

40 × 40 · voxel 31 · 432.9 s · 4 inputs

25 × 25 · voxel 15 · 111.4 s · 4 inputs

40 × 40 · voxel 15 · 75.2 s · 3 inputs

25 × 25 · voxel 25 · 4.8 s · 3 inputs

Top view — 4 inputs, clusters as territories

Top view — 3 inputs

The full map — 1,600 voxel forms from four inputs