diff options
| author | Yuren Hao <yurenh2@illinois.edu> | 2026-07-05 05:12:34 -0500 |
|---|---|---|
| committer | Yuren Hao <yurenh2@illinois.edu> | 2026-07-05 05:12:34 -0500 |
| commit | 5975caf0f52355276aabec5ba8f144db9ac46619 (patch) | |
| tree | c4ee33f44a765ecdd31f0ce57421301bef52d211 /ep_run/ep_c3_watch.py | |
| parent | 75ff326dcb40cd960abd56e9c9c18a45d9e5e2c2 (diff) | |
r-sweep: cos(EP,BPTT) FLAT across r=0.02..0.4 on both operators — nudge-SNR hypothesis refuted
The 50% single-shot sensitivity is MULTIPLICATIVE (near-marginal T2 dynamics'
state-sensitivity; signal and error scale with r together) — not additive
noise divided by 2r. cos ceiling ~0.88 at near-edge states is set by T2
truncation + state sensitivity (0.98 at deeply-contracted states). hr-0.2's
empirical wins are NOT estimator SNR. Hardware upside: algorithm indifferent
to r across 20x -> nudge amplitude free to fight ADDITIVE readout noise.
Surviving accuracy levers: kappa nbrake (Tikhonov, targets the real culprit),
tail-window averaging over argmin selection.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014FAPDWQ49M5Ye3NpTndTpn
Diffstat (limited to 'ep_run/ep_c3_watch.py')
0 files changed, 0 insertions, 0 deletions