From 18c9c1c1d1ec31560c78d41f4ff6957339c1e7c7 Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Wed, 22 Jul 2026 02:31:33 -0500 Subject: experiments: freeze mixed-traffic innovation confirmation --- ROADMAP.md | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'ROADMAP.md') diff --git a/ROADMAP.md b/ROADMAP.md index d6b368d..f8ad2fe 100644 --- a/ROADMAP.md +++ b/ROADMAP.md @@ -41,6 +41,17 @@ mean last-epoch accuracy across top-down traffic scales 0.2 and 0.5 (and the bes endpoint). C1 must use this choice without further test-set tuning. Traffic projection seeds remain separate from model/data seeds and must be recorded explicitly. +The frozen confirmation is MNIST d3/w256 for 15 epochs with the C3-selected low-query calibration +(`K=1`, every four minibatches). It crosses five model seeds with soma traffic at `rho=0.5` and +top-down traffic at `rho=0.2`, each using independently generated projection seeds 1234 and 5678. +Every traffic realization crosses raw, norm-matched raw, neutral-period innovation, and task-period +predictor controls; a no-traffic panel crosses the first three. This is 95 runs total. Test accuracy +is evaluated once after training, while final causal-alignment diagnostics use a fixed 512-example +training prefix. C1 passes only if every one of the four traffic realizations has a mean paired +innovation gain of at least 2 points over norm-matched raw, mean traffic R2 of at least 0.5, and +mean final hidden-layer teaching alignment of at least 0.1. The mean paired no-traffic innovation +change relative to both raw controls must be no worse than -0.5 points. + ### C2. SDIL scales when depth is useful Flattened CIFAR is retained as a preservation result, not as evidence that added depth is used. -- cgit v1.2.3