Decision record: correct distance-physics statement (D_f already D_w-weighted)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
ClaudeandClaude Opus 5.5 committed 2026-11-15 09:06:47 +00:00
1 parent 72ff4fdc4a
commit 0fca206ba9
1 file changed
+2 -2
@@ -14,7 +14,7 @@ Inputs: S2 custody result 1dea89d6, post-hoc D1-D4 daed8e87 (with Mold's final r
- **Irreducible error is unknown.** The labels resolve about 1 % of volume per solver step, and the 2 % non-identifiability stop was withdrawn. How much of R2's 7.3 % can be removed at all has not been measured.
- **The primary metric may not match practical use.** Relative L2 on arrival time weights late, large-T nodes. The model that is best on L2 (R2) is not best on last-filled location (B0). What a molding engineer needs (fill pattern, last-filled region, weld lines, short-shot risk, required pressure) was never written down as the acceptance target.
- **The learning object.** All arms regress T(x) pointwise. Under an approximately constant volumetric flow rate, T(x) = t_fill x (fraction of the cavity volume that fills before x). Time would then be determined almost entirely by the *filling order*, with the scale fixed by physics. B0 is exactly this with graph-distance order, which would explain why it is competitive and why it locates the last-filled region well. **Whether the automatic process in Greif v5 is close to constant flow has not been checked.**
- **No physics in the distance.** R2's graph distance ignores thickness, yet the flow conductance of a thin cavity scales like h^3 (Hele-Shaw). A thickness-weighted (conductance) path distance is a geometry-only quantity we have never tried.
- **Distance physics (CORRECTED 2026-11-15).** My first version said R2's distance ignores thickness. That is wrong: per MeltAstra's source check, the S2 R2/B0 distance D_f is already a tet-edge Dijkstra with edge cost length/mean(D_w), i.e. Gre24's speed f = D_w, and V_f is the full tet-node lumped-volume CDF including d = 0 nodes. What remains untested is (i) a stronger thickness weighting (f = D_w^2, the Hele-Shaw mean-velocity scaling) and (ii) the true-eikonal vs tet-edge-Dijkstra discretisation difference.
- **Scale of the data.** Gre24-style feature models are strong at a few hundred geometries; whole-part neural operators usually rely on far more data. S2 does not tell us whether the neural route fails here because of the data size, the recipe, or something fundamental.
- O1 (type or region organisation with interfaces) is still untested; nothing in S1 or S2 bears on it.
@@ -23,7 +23,7 @@ A. **Change the learning object to filling order, with the time from a volume ma
- Minimal falsifying comparisons, using only existing predictions and the label cache (post-hoc, CPU). The S2 query draws are mass-proportional, so the empirical distribution of the draws estimates F_V over the eligible volume.
- (O-vol oracle) T_hat = c x F_V(rank of the true T). This tests the constant-flow time map itself. If the oracle relative L2 is not well below 7.3 %, route A is falsified as a full-time model: the flow-rate profile matters and the map must be learned. Report c = T_end/t_set fixed from training parts, and the true-T_end variant separately as an oracle.
- (Reorder test) T_hat = c x F_V(rank of R2's prediction), and the same for N and B0. If re-mapping R2's own ordering through the volume map beats R2, the volume-consistent object is better than direct regression without any training.
B. **Physics-structured distance (new).** Compute a conductance-weighted path distance on the tet graph from the existing exact wall distance d(x) (edge cost proportional to 1/h(x)^2 or 1/h^3, with h = 2d as a proxy, chosen before the run). Test it as a one-factor addition to R2 and as the ordering potential for route A, so that geometry carries the flow physics instead of the model having to learn it.
B. **Stronger physics weighting of the distance.** f = D_w is already used (see the correction above). The new test is f = D_w^2, fixed before the run. Test it as a one-factor addition to R2 and as the ordering potential for route A, so that geometry carries the flow physics instead of the model having to learn it.
C. **Hybrid residual network (S3-B, already proposed).** The network corrects R2's prediction, or under route A, the ordering potential. Gate: it must beat R2 by the margin.
D. **Settle the acceptance target with Yanqi.** Agree which outputs matter in practice (last-filled region, weld-line or front-meeting locations, short-shot or pressure risk, full-field T) before choosing between models that rank differently on different metrics.
E. **Data and process (in-house CAE).** Public data cannot test process, gate or material interventions; the bypass pair and process sweeps remain the controlled extension.