feat: add 33 integration proofs verifying architectural claims#101
Merged
Conversation
Implement the Integration Proof Plan v2 with three test suites: - proof_foundation (13 proofs): Berry-Esseen CLT, Fisher sufficiency, XOR self-inverse, triangle inequality, Mexican hat shape, NARS revision monotonicity, ABBA causal retrieval, fusion quality - proof_reasoning_ladder (8 proofs): Parallel error isolation, NARS inconsistency detection, Collapse Gate HOLD/FLOW/BLOCK, thinking style divergence, abduction, counterfactual divergence (Pearl Rung 3), parallel vs sequential probability model - proof_tactics (12 proofs): Recursive expansion, reverse causal trace, adversarial critique, MetaCognition calibration, contradiction detection, CRP distribution, shadow consensus, fusion quality, Hamming Normal approximation, temporal Granger, counterfactual divergence, cross-domain fusion Also includes: - world/counterfactual.rs: Full Pearl Rung 3 do-calculus implementation with intervene(), worlds_differ(), multi_intervene() - src/bin/proof_report.rs: Proof report generator binary - .github/workflows/proof.yml: CI workflow for proof suite - Made core::vsa module public for integration test access All 621 tests pass (587 lib + 33 proof + 1 doctest). https://claude.ai/code/session_01PWXQjdmnMQvZEN5BuBRNMq
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Implement the Integration Proof Plan v2 with three test suites:
proof_foundation (13 proofs): Berry-Esseen CLT, Fisher sufficiency, XOR self-inverse, triangle inequality, Mexican hat shape, NARS revision monotonicity, ABBA causal retrieval, fusion quality
proof_reasoning_ladder (8 proofs): Parallel error isolation, NARS inconsistency detection, Collapse Gate HOLD/FLOW/BLOCK, thinking style divergence, abduction, counterfactual divergence (Pearl Rung 3), parallel vs sequential probability model
proof_tactics (12 proofs): Recursive expansion, reverse causal trace, adversarial critique, MetaCognition calibration, contradiction detection, CRP distribution, shadow consensus, fusion quality, Hamming Normal approximation, temporal Granger, counterfactual divergence, cross-domain fusion
Also includes:
All 621 tests pass (587 lib + 33 proof + 1 doctest).
https://claude.ai/code/session_01PWXQjdmnMQvZEN5BuBRNMq