Giorgio Gilestro
6f8cef1ac5
Removed from main (all preserved on the dev branch): the arXiv build and
its sources, design documents (blueprint, results summary, review responses,
essay drafts), tasks/ and CLAUDE.md, the cover letter and reference tooling,
two unused manuscript figures, and every experiment that feeds no figure or
number in the paper: the collapse null, the sexual-vs-asexual lineage, the
NK speciation variant, the 0.5B single-seed LLM prototypes, the compose and
society experiments with their calibration and pilot runs, and their
configs, runners, tests, figure scripts and PBS jobs. Their result bundles
are moved to results/_archive/ (ignored) so the parquets stay on disk.
Also: plot_llm_speciation reads the s{seed}/ layout; the mating-breadth
plot writes under its bundle name; Makefile targets reduced to the kept
experiments; REPRODUCING.md and README point to dev for the rest.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y64o8FKP7rCuXzC48pxpMm
|
||
|---|---|---|
| .. | ||
| manifest.json | ||
| README.md | ||
| recombination.pdf | ||
| recombination.png | ||
| resolved_config.yaml | ||
recombination — "merge, don't average" holds in real neural weights (E4 in the flesh)
Claim tested: the E4 finding — complementary specialists can rebuild the rare tail, but only if you merge them rather than average them — was proven in math. Does it survive in trained neural nets?
Setup (Layer 1.5). K_T specialist GRUs are each trained on a different slice of the rare
tail (slices set by the exact shared-switch construction, so teacher count K_T, correlation ρ and
retention q stay clean knobs and the construction-level union matches the closed form). The pupil then
recombines the teachers' measured distributions two ways: mean (naive pooling) vs
oracle-guided max-merge (per mode, keep the strongest teacher — a union-preserving merge). K = 256,
n = 200, q = 0.5, swept K_T ∈ {1,2,3,5} × ρ ∈ {0, 1}, 8 repeats.
Symbols
K_Tnumber of specialist teachers;ρhow correlated their retained tails are (0 = complementary, 1 = identical clones).- union coverage — tail covered by ≥1 teacher (the raw supply). surviving coverage — what remains after the pupil resamples.
- mean-distill — average/pool the teachers. max-merge — keep each mode's strongest teacher (union-preserving).
- target vs trained — coverage computed from the assigned distributions vs from the trained GRU outputs.
The four panels
- Supply. Union coverage vs
K_T: atρ = 0it climbs 0.49 → 0.96 and matches the closed formU(K_T, ρ, q); atρ = 1(clones) it is flat. Diverse teachers supply more tail. - Analytic teachers (
ρ = 0). Surviving coverage vsK_T: max-merge (green) rises 0.043 → 0.087 while mean-distill (red) stays flat ~0.045 — the conservation law from E4, reproduced on the assigned distributions. - Trained GRU teachers (
ρ = 0). The same comparison on real trained weights: same signs — max-merge above mean-distill — but compressed and noisier (the GRU's smoothing inflates the baseline and the deep tail barely clearsn = 200resampling). The honest inductive-bias caveat. - Control (
ρ = 1). Identical teachers: union, max and mean are all flat — more clones buy nothing. Decorrelation is what the benefit needs.
Takeaway
"Merge, don't average" is not a mathematical artefact — in trained neural nets, a union-preserving
max-merge realises the multi-teacher tail benefit while naive averaging conserves the collapse.
This is load-bearing for the paper's recombination claim (and points at merging over distillation for
the deferred LLM layer). Falsifier (not triggered): if mean-distill had also risen with K_T, or
max never beat mean, the recombination lesson would have died in real weights.