Run · Voynich Manuscript
Generation 5: population annealing from a higher starting beta (0.6); method dropped after two generations
Population-annealed EM for the lattice key, compared with independent restarts: Training log-likelihood and key accuracy on the one-sign-per-letter control, against independent restarts.
Result
| Run | Measured | Result | Against | Decision | Date |
|---|---|---|---|---|---|
| Run | Training log-likelihood and key accuracy on the one-sign-per-letter control, against independent restarts | ll -41,455 (all 32 members converge to one state), key 0.218, coverage 0.328 | 200 independent restarts, same control: ll -40,969, key 0.698 | Population annealing loses to independent restarts on the planted control in both generations (beta from 0.2: ll -42,067, key 0; from 0.6: ll -41,455, key 0.22). Resampling drives the population onto one state, which removes the diversity that restarts depend on. Dropped, with deterministic annealing. |
What it used and found
Latin genre texts
DATDataset
Parameters
PA = 8, pw = 0.25, BETA0 = 0.6, alpha = 0.05, population = 32, permutations = 10, em_iterations = 60
Cite it
Mysteries Department. (2026). Generation 5: population annealing from a higher starting beta (0.6); method dropped after two generations [Run RUN0DCGYKWPD5D2]. https://mysteriesdept.com/r/RUN0DCGYKWPD5D2
@misc{mysteriesdept_run0dcgykwpd5d2,
title = {Generation 5: population annealing from a higher starting beta (0.6); method dropped after two generations},
author = {{Mysteries Department}},
year = {2026},
note = {Run, catalog key RUN0DCGYKWPD5D2},
url = {https://mysteriesdept.com/r/RUN0DCGYKWPD5D2}
}