diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-07-22 13:33:33 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-07-22 13:33:33 -0500 |
| commit | 0db651ac83608ea8f5b94950a391da22f076a142 (patch) | |
| tree | aae335ea16eb1b2f8981ba1217b2f91cdc2ae9b9 /README.md | |
| parent | c27e0824a41e6c54dd9079f17c20c306c5f2e9c9 (diff) | |
baseline: add normalized local response mirroring
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 9 |
1 files changed, 9 insertions, 0 deletions
@@ -130,6 +130,15 @@ child/parent correlation. Its BatchNorm-coupled JVP and symmetric-limit delta rule are executable smoke tests. `ORAL_A_V4.md` freezes the causal-capture and conditional accuracy gates; no accuracy claim is attached until they pass. +Because the global-scalar V4 gate failed, the next strong inherited comparator +uses normalized local response mirroring. Its observation phase sends Gaussian +parent activity through ordinary local forward synapses; a separate feedback +update sees only the probe/child-response pairs. The smoke suite verifies both +high finite-sample kernel recovery and that changing forward parameters after +observation cannot affect the update. Any scale obtained this way is credited +to the weight-mirror baseline until innovation residualization is shown to be +load-bearing on top of it. + The subsequent V3 mechanism estimates the required A/G matrix statistics directly by perturbing the vectorizer parameter subspace. It remains forward-only and uses two causal loss queries per direction. Its exact and |
