logaritmisk/trueskill-tt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

docs: fix stale numerics in t4-margin-factor plan

Browse Source 
The plan's prose quoted Z_cav ≈ 0.046827 and log_evidence ≈ -3.0613,
which diverged from the values asserted by the shipped test in
src/factor/mod.rs (-3.062235327364623). Update prose and the matching
code comment to 0.04678 / -3.0622.
This commit is contained in:
Anders Olsson
2026-05-08 14:37:58 +02:00
parent f6a83e4dc6
commit 1445c08896
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/superpowers/plans/2026-04-27-t4-margin-factor.md
+2 -2
View File
@@ -49,7 +49,7 @@ A Gaussian `N(m, σ)` constructed via `Gaussian::from_ms(m, σ)`. Multiplication
**Concrete numerical check for tests:** With cavity `N(0, 6)` and observation `m_obs=5, σ=1`: **Concrete numerical check for tests:** With cavity `N(0, 6)` and observation `m_obs=5, σ=1`:
- `D_cav.pi = 1/36 ≈ 0.027778`, `D_cav.tau = 0`. - `D_cav.pi = 1/36 ≈ 0.027778`, `D_cav.tau = 0`.
- New marginal: `pi = 0.027778 + 1 = 1.027778`, `tau = 0 + 5 = 5`. So `mu = 5 / 1.027778 ≈ 4.864865`, `sigma = 1/sqrt(1.027778) ≈ 0.986394`. - New marginal: `pi = 0.027778 + 1 = 1.027778`, `tau = 0 + 5 = 5`. So `mu = 5 / 1.027778 ≈ 4.864865`, `sigma = 1/sqrt(1.027778) ≈ 0.986394`.
- `Z_cav = pdf(5, 0, sqrt(36 + 1)) = pdf(5, 0, sqrt(37)) ≈ 0.046827`. So `log_evidence ≈ -3.0613`. - `Z_cav = pdf(5, 0, sqrt(36 + 1)) = pdf(5, 0, sqrt(37)) ≈ 0.04678`. So `log_evidence ≈ -3.0622`.
--- ---
@@ -182,7 +182,7 @@ mod tests {
f.propagate(&mut vars); f.propagate(&mut vars);
let z = f.evidence_cached.unwrap(); let z = f.evidence_cached.unwrap();
// pdf(5, 0, sqrt(37)) ≈ 0.046827 // pdf(5, 0, sqrt(37)) ≈ 0.04678
assert!((z - 0.04682752233851171).abs() < 1e-10); assert!((z - 0.04682752233851171).abs() < 1e-10);
// Subsequent propagations don't change it. // Subsequent propagations don't change it.