# Эксперимент: Unassembled Adaptive Cell Neural Network (#1288)

{
  "worked": false,
  "confidence": 8,
  "verdict": "Built an unassembled cell-tensor network with segmented valence-weighted synchronization and a matched assembled message-passing baseline. The core projection math was verified exactly: fixation, idempotence, symmetry, and segmented reduction all had zero numerical error. Delayed synchronization showed a promising loss/throughput tradeoff, but it did not establish the proposed 1.5x speedup or memory reduction, so this toy experiment does not justify claiming a practical win.",
  "metrics": {
    "baseline": "assembled: validation MSE 0.15475, 6484.2 examples/s, 1.579 s",
    "idea": "cell_sync_1: MSE 0.01258, 4230.7 examples/s; cell_sync_4: MSE 0.00874, 4374.2 examples/s; cell_sync_8: MSE 0.02608, 7503.1 examples/s. Projection errors were 0.0 for fixation, idempotence, symmetry, and manual segmented averaging; pre-sync RMS disagreement was 0.8176."
  },
  "how_to_run": "/home/maxwelhelp/main/bin/python3 experiment.py",
  "files": [
    "experiment.py",
    "results.json"
  ],
  "limitations": "Only a regular 8x8 quad mesh and synthetic neighbor-averaging regression were tested; no adaptive quadtree, Poisson coefficient-to-solution task, hierarchy transfer, peak-memory measurement, multi-seed statistics, or larger-scale sparse assembled implementation was evaluated. The toy cell and assembled models also have different parameterizations, so throughput is indicative rather than a rigorous apples-to-apples benchmark."
}