diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-03-24 20:07:03 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-03-24 20:07:03 -0500 |
| commit | 825d973428450cb24d8cccc8c2604235ef974b7c (patch) | |
| tree | 865bf6f7cc5eabbdbbccfb5c14c927584dd1a4f8 /report/cifar_feature_drift.png | |
| parent | 5550e2cac45758e579810ae36bf716a0b819cebc (diff) | |
Add Phase 6: snapshot exploitability reveals local update rule is the bottleneck
Phase 6A: Better credit is ANTI-CORRELATED with loss decrease on fixed snapshot.
DFA (Gamma=0.01) → dL=-0.0001 (only method that decreases loss)
Vec_M4 (Gamma=0.38) → dL=+0.057 (increases loss most)
Oracle BP (Gamma=1.0) → dL=+0.011 (still increases loss)
Phase 6C: Target-shift rule reduces damage but cannot make non-DFA credits productive.
The inner-product surrogate <F_l(h), a_l> is fundamentally mismatched with directional credit.
Conclusion: Case B — the primary bottleneck is the local update paradigm itself,
not the credit estimator quality or tracking/co-adaptation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Diffstat (limited to 'report/cifar_feature_drift.png')
0 files changed, 0 insertions, 0 deletions