summaryrefslogtreecommitdiff
path: root/sdil
AgeCommit message (Collapse)Author
2026-08-29feat: add correlated autozero samplingYurenHao0426
2026-08-29feat: add hardware autozero SDIL primitiveYurenHao0426
2026-08-29feat: match physical classification early stoppingYurenHao0426
2026-08-29feat: train the reconstructed physical gridYurenHao0426
2026-08-29feat: reconstruct the Dillavou physical gridYurenHao0426
2026-08-29feat: model state-dependent physical learning biasYurenHao0426
2026-08-29feat: compose SDIL with physical overclampingYurenHao0426
2026-08-10exp: freeze BabyAI population predictor screenYurenHao0426
2026-08-10feat: add basal history to BabyAI policyYurenHao0426
2026-08-10exp: freeze BabyAI shared-feedback protocolYurenHao0426
2026-08-10feat: add local online least-squares residual predictorYurenHao0426
2026-08-07feat: transfer released physical drift profile to Rain EPYurenHao0426
2026-08-07fix: freeze constant update calibration after one probeYurenHao0426
2026-08-07feat: add Dillavou EP update-bias protocolYurenHao0426
2026-08-06feat: make Rain hardware bias beta-independentYurenHao0426
2026-08-06feat: add BP-free Rain layer-state adapterYurenHao0426
2026-08-06feat: add per-cell batched local predictorYurenHao0426
2026-08-06feat: support frozen Rain neutral predictorsYurenHao0426
2026-08-06feat: report Rain bias-to-signal ratiosYurenHao0426
2026-08-06feat: add counted Rain neutral calibrationYurenHao0426
2026-08-06feat: integrate local debiaser with Rain EPYurenHao0426
2026-08-06feat: add shared no-grad two-state debiaserYurenHao0426
2026-08-06results: test SDIL on measured physical surrogateYurenHao0426
2026-08-06feat: add BP-free physical coupled-learning adapterYurenHao0426
2026-08-06experiment: implement causally whitened feedback fitsYurenHao0426
2026-08-06experiment: implement layerwise causal feedback bootstrapYurenHao0426
2026-08-06experiment: implement shared-feedback feasibility gateYurenHao0426
2026-07-30experiment: package portable A6000 crossoverYurenHao0426
2026-07-27baseline: add Transformer Dual Prop and EPYurenHao0426
2026-07-27baseline: add matched Transformer Forward-ForwardYurenHao0426
2026-07-27baseline: add matched Transformer PEPITAYurenHao0426
2026-07-27baseline: add strict Transformer DFAYurenHao0426
2026-07-27baseline: add matched Transformer feedback transportYurenHao0426
2026-07-27baseline: add matched ResNet equilibrium propagationYurenHao0426
2026-07-27baseline: add matched ResNet Forward-ForwardYurenHao0426
2026-07-27baseline: add matched ResNet PEPITAYurenHao0426
2026-07-27baseline: add matched ResNet Dual PropagationYurenHao0426
2026-07-26experiment: freeze KP teaching-signal ablationYurenHao0426
2026-07-23protocol: freeze oral-B-v2 cold-start recoveryYurenHao0426
2026-07-23bci: audit v2 outcome-surprise signaturesYurenHao0426
2026-07-23bci: add endpoint-free oral-B-v2 mechanicsYurenHao0426
2026-07-22audit: isolate BCI role learner from environment mapYurenHao0426
2026-07-22mechanism: separate BCI role from performance velocityYurenHao0426
2026-07-22experiment: support audited dynamic projection endpointsYurenHao0426
2026-07-22experiment: add dynamic neutral residual projectionYurenHao0426
2026-07-22experiment: add one-sided residual stability marginYurenHao0426
2026-07-22experiment: freeze predictor after neutral fitYurenHao0426
2026-07-22experiment: add closed-form neutral innovation fitYurenHao0426
2026-07-22audit: conservatively count predictor arithmeticYurenHao0426
2026-07-22audit: align mixed-traffic compute with cost ledgerYurenHao0426
generated by cgit v1.2.3 (git 2.25.1) at 2026-09-15 19:29:45 +0000