Skip to content

Withdraw BETSE analysis report; correct validation claims - #2

Merged
bshepp merged 1 commit into
mainfrom
withdraw-betse-report
Sep 29, 2026
Merged

bshepp merged 1 commit into
mainfrom
withdraw-betse-report

Conversation

@bshepp

@bshepp bshepp commented Sep 27, 2026

Copy link
Copy Markdown
Owner

Summary

Withdraws the public BETSE analysis report and corrects claims in the README and docs index.

Why

A review found that the report's conclusions do not hold:

  • The BETSE loader sorted exported frames incorrectly, so every time-dependent result was computed on a scrambled sequence.
  • The Lyapunov values came from the estimator removed in Tier 0.
  • No claim was tested against a null model.

Changes

  • docs/BETSE_ANALYSIS_REPORT.md is replaced with a withdrawal notice. The original stays in git history.
  • The report is removed from the site navigation and the docs index.
  • The README no longer states that the tool never reports chaos without surrogate evidence. That is true of classify_attractor() but not of the pipeline or the mneme.core.attractors detectors.
  • The docs index no longer advertises the Wolf algorithm.

Code fixes for the underlying defects follow in a separate PR.

🤖 Generated with Claude Code

The report's time-dependent results were computed on frames loaded out of
time order, and its Lyapunov values came from the removed estimator.
Replace the report with a withdrawal notice, remove it from the site
navigation, and correct the README and docs index claims about chaos
gating and the Wolf algorithm.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
@bshepp
bshepp merged commit c1eabf6 into main Sep 29, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant