sdil.git - Unnamed repository; edit this file 'description' to name the repository.

Age	Commit message (Expand)	Author
10 days	oral-a: add audited convolutional runner	YurenHao0426
10 days	oral-a: add convolutional apical calibration	YurenHao0426
10 days	oral-a: add exact local convolution eligibilities	YurenHao0426
10 days	oral-a: add deterministic CIFAR image pipeline	YurenHao0426
10 days	results: close oral B signature screen	YurenHao0426
10 days	experiments: freeze oral B development screen	YurenHao0426
10 days	bci: add paired phase-lesion runner	YurenHao0426
10 days	bci: add preregistered signature metrics	YurenHao0426
10 days	bci: add continuous local-learning dynamics	YurenHao0426
10 days	experiments: preregister Harnett signature gates	YurenHao0426
10 days	baselines: freeze strict C4 audit gate	YurenHao0426
10 days	baselines: audit native author runs	YurenHao0426
10 days	figures: make audited PDFs deterministic	YurenHao0426
10 days	docs: freeze native baseline reproductions	YurenHao0426
10 days	results: close normalized C2 recovery	YurenHao0426
10 days	experiments: freeze normalized C2 recovery	YurenHao0426
10 days	sdil: normalize causal vectorizer regression	YurenHao0426
10 days	theory: prove estimator and innovation regimes	YurenHao0426
10 days	results: close feedback-first C2 recovery	YurenHao0426
10 days	experiments: freeze feedback-first C2 screen	YurenHao0426
10 days	sdil: add feedback-first apical calibration	YurenHao0426
10 days	results: close calibration quality recovery	YurenHao0426
10 days	experiments: freeze C2 calibration quality recovery	YurenHao0426
10 days	results: confirm C2 feedback amortization bottleneck	YurenHao0426
10 days	experiments: freeze direct perturbation C2 diagnosis	YurenHao0426
10 days	baseline: add unamortized node perturbation	YurenHao0426
10 days	results: close context-vectorizer C2 recovery	YurenHao0426
10 days	results: retain failed context-vectorizer recovery	YurenHao0426
10 days	experiments: freeze context-vectorizer C2 recovery	YurenHao0426
10 days	fix: bound state-conditioned feedback features	YurenHao0426
10 days	method: condition apical vectorizers on neural state	YurenHao0426
10 days	results: close failed C2 local panel	YurenHao0426
10 days	results: retain failed useful-depth local panel	YurenHao0426
10 days	experiments: freeze useful-depth local-rule panel	YurenHao0426
10 days	results: mark useful-depth BP screen passed	YurenHao0426
10 days	results: establish BP useful-depth candidate	YurenHao0426
10 days	experiments: freeze useful-depth BP screen	YurenHao0426
10 days	experiments: audit residual block usage with lesions	YurenHao0426
10 days	results: close failed C1 recovery branch	YurenHao0426
10 days	results: retain failed FashionMNIST C1 recovery	YurenHao0426
10 days	experiments: freeze FashionMNIST C1 recovery	YurenHao0426
10 days	results: record frozen C1 near-miss	YurenHao0426
10 days	results: confirm mixed-traffic innovation across seeds	YurenHao0426
10 days	experiments: freeze mixed-traffic innovation confirmation	YurenHao0426
10 days	experiments: isolate final diagnostics from held-out data	YurenHao0426
10 days	results: mark amortized calibration gate passed	YurenHao0426
10 days	results: confirm low-query calibration across seeds	YurenHao0426
10 days	experiments: freeze five-seed low-query confirmation	YurenHao0426
10 days	experiments: log peak memory and total training work	YurenHao0426
10 days	results: freeze single-direction calibration	YurenHao0426