index
:
sdil.git
a6000-x1-results
main
Unnamed repository; edit this file 'description' to name the repository.
Ubuntu
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
experiments
Age
Commit message (
Expand
)
Author
2026-08-29
theory: connect local bias removal to CLLN scaling
YurenHao0426
2026-08-29
test: audit CLLN locality and numerical invariants
YurenHao0426
2026-08-29
analysis: support sharded CLLN confirmations
YurenHao0426
2026-08-29
results: freeze size-matched overclamp selector
YurenHao0426
2026-08-29
analysis: add CLLN scaling pilot figure
YurenHao0426
2026-08-29
feat: add clean-only overclamp selection path
YurenHao0426
2026-08-29
feat: add stable-target and physical-exposure metrics
YurenHao0426
2026-08-29
feat: add sparse digital CLLN scaling ladder
YurenHao0426
2026-08-29
analysis: add clustered correlated autozero intervals
YurenHao0426
2026-08-29
exp: add correlated autozero robustness matrix
YurenHao0426
2026-08-29
docs: label SPICE reference configuration
YurenHao0426
2026-08-29
fix: separate autozero SPICE switching phases
YurenHao0426
2026-08-29
exp: add SPICE autozero primitive check
YurenHao0426
2026-08-29
analysis: add clustered autozero intervals
YurenHao0426
2026-08-29
exp: add hardware autozero robustness matrix
YurenHao0426
2026-08-29
fix: count divergent physical calibration trials
YurenHao0426
2026-08-29
exp: stress test SDIL calibration
YurenHao0426
2026-08-29
fig: summarize the full physical grid crossover
YurenHao0426
2026-08-29
fig: plot physical grid bias crossover
YurenHao0426
2026-08-29
exp: add physical grid bias crossover
YurenHao0426
2026-08-29
exp: add physical grid clean-learning gate
YurenHao0426
2026-08-29
feat: train the reconstructed physical grid
YurenHao0426
2026-08-29
data: extract the released Figure 5 protocol
YurenHao0426
2026-08-29
feat: reconstruct the Dillavou physical grid
YurenHao0426
2026-08-29
results: validate SDIL on device imperfection model
YurenHao0426
2026-08-29
exp: add device-level imperfection crossover
YurenHao0426
2026-08-29
feat: model state-dependent physical learning bias
YurenHao0426
2026-08-29
exp: add physical clamp budget crossover
YurenHao0426
2026-08-29
feat: compose SDIL with physical overclamping
YurenHao0426
2026-08-27
exp: freeze raw KP traffic scaling control
YurenHao0426
2026-08-10
results: close BabyAI population predictor rescue
YurenHao0426
2026-08-10
exp: freeze BabyAI population predictor screen
YurenHao0426
2026-08-10
results: reject diagonal SDIL on PickupLoc
YurenHao0426
2026-08-10
results: select four-step PickupLoc history
YurenHao0426
2026-08-10
feat: add basal history to BabyAI policy
YurenHao0426
2026-08-10
results: close feedforward PickupLoc gate
YurenHao0426
2026-08-10
results: retain GoToObj as diagnostic only
YurenHao0426
2026-08-10
results: select BabyAI clean configuration
YurenHao0426
2026-08-10
fix: resolve visible CUDA device for BabyAI
YurenHao0426
2026-08-10
exp: freeze BabyAI shared-feedback protocol
YurenHao0426
2026-08-10
fix: separate template train and test instructions
YurenHao0426
2026-08-10
exp: freeze Larkum public innovation pilot
YurenHao0426
2026-08-10
exp: confirm four-probe SDIL across seeds
YurenHao0426
2026-08-10
exp: parameterize upfront predictor screen
YurenHao0426
2026-08-10
feat: add local online least-squares residual predictor
YurenHao0426
2026-08-10
exp: compare equal upfront calibration budgets
YurenHao0426
2026-08-10
fix: honor Dillavou upfront calibration budget
YurenHao0426
2026-08-08
exp: parameterize full EP confirmation seeds
YurenHao0426
2026-08-07
fix: summarize Rain metric trajectories
YurenHao0426
2026-08-07
exp: test state-aware correction on full EP trajectories
YurenHao0426
[next]