Asking the fleet what it is doing…
monad-knowledge Wi-Fi sensing lab · FIIT STU
Campaign session

01KWYBEJQK4307TTP9KMRY0DEY

finished 2026-07-07 13:14:33.075446+00:00 → 2026-07-07 14:03:20.367105+00:00 · 90 runs · supervisor: react-agent

“Phase-1 pilot (6 of 10 floors staged): machinery validated + variance-split settled, powered primary not met as pre-declared. LOFO cross-floor inflation pooled ×3.37 (honest leave-one-occupancy-out within 0.026 vs held-out 0.088), floor-cluster 95% CI [1.04, 9.85] — at 6 clusters the interval is unreliable, its exclusion of 1.0 not leaned on. The ×3.37 hides extreme per-floor heterogeneity (0.70× resplan-7421 no-inflation → 49.8× resplan-1374; mean-of-ratios 10.1 vs ratio-of-means 3.37) — that spread IS why the CI is order-of-magnitude wide. C2 geometry-dominates the CV(N) signal DECISIVELY (σ²_floor 0.0109 vs σ²_seed 0.000165, 66×; between-floor SD 0.105 vs between-seed 0.0116). C3 honest held-out baseline used. Criteria 2/3 (C2, C3; C1 under-powered fail). geometry-dominates is a CV(N) property, distinct from the MAE-inflation drift. No hypothesis strength change; design input quantifying that Phase 2 needs ≥40 floors. 90 runs gate-passed.”

Archive snapshot, as of 1 day ago — the run corpus is rebuilt once a day, so this page is not a live reading. The fleet panel is the live one; it refreshes every 30 s.

Success criteria

CriterionResolved
Primary: bootstrap-over-floors 95% CI on LOFO inflation ratio lower bound > 1.5×. — FAIL: CI [1.04, 9.85], lower bound 1.04 < 1.5; at 6 clusters the interval is unreliable. Pooled ×3.37 is a heterogeneous mixture (0.70×–49.8× per floor). Under-powered by design (n0≈40 floors). no
Variance split: σ²_floor and σ²_seed reported; geometry-dominates iff between-floor SD of per-floor CV(N) > within-floor between-seed SD. — PASS: σ²_floor 0.0109 vs σ²_seed 0.000165 (66×); between-floor SD 0.105 vs between-seed SD 0.0116 (9×). Robust at n=6/3. Measured on CV(N). yes
Comparator honesty: LOFO denominator is held-out within-floor leave-one-occupancy-out, never in-sample residual. — PASS: leave-one-occupancy-out within (0.026) vs held-out (0.088); generalization error, fixing the original in-sample-residual bug. yes

Synthesis

c-csi-cross-geometry-scaleout — Phase 1 pilot, session 01KWYBEJQK4307TTP9KMRY0DEY (2026-07-07)

Verdict: the Phase-1 pilot validates the machinery and settles the variance split — geometry dominates seed decisively in the CV(N) sensing signal — but the powered primary criterion is not met, and at 6 floors the inflation-ratio CI is too fragile to reason over. LOFO cross-floor inflation is a pooled ×3.37 (honest leave-one-occupancy-out within-floor baseline 0.026 vs held-out 0.088), floor-cluster 95% CI [1.04, 9.85] — but with only 6 bootstrap clusters that interval has erratic coverage; its bare exclusion of 1.0 is not trustworthy evidence and is not leaned on. Criteria 2/3: C2 (variance split) and C3 (honest baseline) pass; C1 (inflation CI lower bound > 1.5) fails on width.

What we ran

6 ResPlan floors × 3 seeds × 5 occupancy levels {0,2,4,6,8} at 2.4 GHz, 48 Monte-Carlo body placements per cell — 90 exp-csi-static runs, all gate-passed, local Sionna, ble disabled (CSI-only estimand). Scope cap disclosed up front: the brief's Phase-1 names 10 floors, but only 6 are staged in PostGIS (resplan-{12439,1374,147440,7421,16157,12419}; the other four — 10156/10425/6090/8266 — have no experiment). n₀ = z²σ²/d² ≈ 40 floors for the powered primary, so 6 floors is a machinery-and-variance pilot, not the powered test. Reduction csi_cross_geometry_scaleout.py (unit = floor).

What we found

  1. Geometry dominates seed — decisively (C2 ✓). Nested one-way random-effects decomposition on the per-(floor,seed) mean-CV signature: σ²_floor = 0.0109 vs σ²_seed = 0.000165 — floor variance is 66× the seed variance. Between-floor SD 0.105 vs within-floor between-seed SD 0.0116 (9×). The CV(N) curve is a stable floor signature; placement-seed jitter is small against the geometry spread. This is the pilot's real deliverable: the cross-floor drift the thesis chain leans on is a genuine geometry effect, not seed noise.
  2. Honest baseline changes the comparator, not the direction (C3 ✓). The original csi_cross_geometry.py used an in-sample fit residual as the "within-floor" MAE (the audit's flagged bug). This reduction uses leave-one-occupancy-out within each floor (0.026) as the denominator — a generalization error, never the training residual. LOFO held-out MAE 0.088 → inflation ×3.37.
  3. Primary CI is wide and its width is the finding (C1 ✗, by design). Floor-cluster bootstrap (unit = floor, 6 clusters) gives inflation 95% CI [1.04, 9.85]. The pooled ×3.37 hides extreme per-floor heterogeneity — the per-floor held-out/within ratios span 0.70× (resplan-7421, where held-out is actually better than within — no inflation) to 49.8× (resplan-1374, whose tiny within-MAE 0.005 dominates the pooled numerator); mean-of-ratios is 10.1 vs ratio-of-means 3.37. That spread, not sampling luck, is why the CI is nearly an order of magnitude wide. At 6 clusters the bootstrap coverage is itself unreliable, so the interval is best read as "uninformative but consistent with large inflation," not as evidence the lower bound clears 1.0. Booked false, not rounded up.

What it means for the thesis chain

The CV(N) sensing signal is geometry-dominated (σ²_floor ≫ σ²_seed) — that variance-split result is solid at this n and is the pilot's real contribution. It is distinct from, and does not by itself establish, the MAE-inflation drift claim (C1 is on held-out MAE, C2 on CV(N) variance — related but not the same quantity). The honest inflation point estimate is ×3.37, but it is a heterogeneous mixture (0.70×–49.8× across floors) and the number the thesis leans on cannot carry a > 1.5 lower-bound claim until Phase 2 stages ≥40 floors (the ResPlan converter → gis_rebuild_nrggis_validate batch). No hypothesis strength change: this is a design input that quantifies how many distinct geometries the IP-106 field study needs, and the answer is "many — seed replication is cheap, floor replication is what buys the CI." Phase 2 is the powered follow-on.

Honest scope

In-silico, ResPlan geometries, single band (2.4 GHz), CSI-only (ble disabled). 6 of 10 planned floors, 3 seeds each. The variance-split and honest-baseline results are solid at this n; the inflation-ratio CI is deliberately under-powered and labelled so. Reduction + cross_geometry.parquet (per floor×seed×N) + per-floor CSV + 2 figures (csi_cross_geometry small-multiples with the CI panel, seed_trace between-floor vs between-seed spread) under this session's artefacts/ prefix; 90 runs attached.

Criticism adversarial review

Written by the campaign-critic subagent against the brief's success criteria — read it as the counter-position to the synthesis above.

campaign-critic — c-csi-cross-geometry-scaleout / 01KWYBEJQK4307TTP9KMRY0DEY

Agrees with synthesis. Severity: medium. Verdict sound, C1 correctly booked fail. TWO framing fixes integrated: (1) "confirmed" over-claimed the failing C1 CI — reframed as uninformative at 6 clusters. (2) ×3.37 hid per-floor heterogeneity (0.70× resplan-7421 no-inflation → 49.8× resplan-1374; mean-of-ratios 10.1 vs ratio-of-means 3.37) — surfaced as the real finding. (3) C2 (CV(N) variance) separated from C1 (MAE inflation). C2 (66×/9× gap, robust at n=6) and C3 (honest leave-one-occupancy-out) genuinely supported. Scope: 6-floor/3-seed in-silico pilot, under-powered for C1 by design, bootstrap bounds unreliable at 6 clusters; settles only that machinery gate-passes and geometry ≫ seed in CV(N).

Attached runs

Run Gate Purpose Replay
2AHTREW9
E995S9K6
MBW1NF3M
5XSBAFWE
JF4XPBB6
KDB81FPQ
3RH8QYG3
TG9P5T40
0M19D6QN
QS4NJ0Z8
KSRAZFY6
A7CEEAFN
18K7V157
GJ614BHE
G04NJ5FF
ZP9SV7JB
38VJ4X4F
595DBD81
ANY5MRE4
MC16JZBN
KZR2S07S
7QPMNBQ4
39SJ3XX2
4J0MJJ5H
T2K2GX2Y
MESSF2XH
AG366MSR
AY759EP2
8YRDBXX6
Q8KEHJ07
1AQ115XQ
BTYS6BQ1
FJF648DQ
90F67KEW
464N6Z1K
Q0NHR8DE
0NQN2V0C
Z21G7FF4
NBPH0SBR
36ZV8WFH
E4KG1ZKY
ZQ13BZZ5
9GJW4EN8
JE3QFNN0
TPBTCZQG
NPGVF6CE
J1FYNF6B
FVBERNZ1
VPJ8Y5WM
7N4R3K42
MJQPHRAN
DPTW96ZH
04FR6Z2E
F17N80TF
9CSCH5TP
S9YZ609X
DKG3ECGT
8CEQYMT8
1544GYPG
V6YX0Z97
8H3QRVPK
F7J4YTCG
GB111029
H42FF399
EYCEX7T7
40TKFR1G
CWJRK1ZC
Q2K19FRA
4F846MF3
JX1QCNQ8
D7T0283Y
TYVRWE1R
B7BV2BDM
5NDWT2C2
GQ2X9H39
RTRATWRQ
PA63K627
C6AGJVY5
H4HGHMS2
4FDDS1S6
CCP4AHVG
634B23QS
AE5BXPAW
DWHJ86EG
MFDN2J0C
SES266BD
DWFDXFKN
6TJFAN1Q
4NNF1K3N
KT2NPFXY