results: compare full state-bias EP trajectories

1 files changed, 22 insertions, 0 deletions

diff --git a/RAIN_EP_RELEASED_PROFILE.md b/RAIN_EP_RELEASED_PROFILE.md
index eb6a16f..6849a9d 100644
--- a/RAIN_EP_RELEASED_PROFILE.md
+++ b/RAIN_EP_RELEASED_PROFILE.md
@@ -123,3 +123,25 @@ Affine SDIL lost all four paired comparisons. The one-epoch state-dependent
 claim is therefore negative. The remaining accuracy question is whether the
 affine predictor needs a longer trajectory to learn the state relation; R3
 tests that question on full-data ten-epoch trajectories.
+
+## R3 full-data trajectories
+
+R3 used all 60,000 FashionMNIST training examples, the official 10,000-example
+test set, ten epochs, and two paired initializations. The hardware bias was the
+released state-dependent profile, not the constant-offset sanity check. The
+intercept tracker and affine SDIL received the same exact neutral observation
+every ten parameter updates. Final test accuracy was:
+
+| seed | clean/oracle | raw | intercept tracker | affine SDIL | SDIL minus intercept |
+|---:|---:|---:|---:|---:|---:|
+| 1988 | 88.67% | 10.00% (nonfinite after epoch 3) | 84.52% | 83.13% | -1.39 points |
+| 1989 | 88.80% | 10.00% (nonfinite after epoch 2) | 86.02% | 87.22% | +1.20 points |
+| mean | 88.74% | 10.00% | 85.27% | 85.18% | -0.10 points |
+
+Both local correction methods prevent the raw state-dependent failure, but
+affine SDIL does not have a consistent accuracy advantage: it wins one pair,
+loses one pair, and is 0.095 percentage points lower on the two-seed mean. The
+predeclared endpoint is final accuracy, so its temporarily higher intermediate
+or best accuracy is not substituted for this result. Two further paired seeds
+are required to narrow the comparison, but R3 itself is mixed/negative for an
+SDIL-over-intercept claim.