A new fit of voltage history weights reduced state error after map clipping in nine SEEG rest records with sparse simulated loss
We tested an early ridge fit with a 0.95 norm limit on the total gap state map. The test used nine public SEEG rest records and sparse simulated input loss. The constrained fit reduced normalized state error from 1.232 to 0.514 against clipping after fitting. Error was lower in all nine records. Fitting only the voltage history weights at the same clipped map gave almost the same result. Joint map fitting gave no further test gain. Dense loss gave lower state error after either new fit, but fixed tight timing gave no phase gain against clipping alone. All 144 fits passed numerical checks. Fresh raw prefixes reproduced the selected voltage data exactly. The package includes data, source evidence, code, models, figures, a proof, and repeat instructions. This exploratory test reused selected records. It did not test real stimulation artifacts or clinical performance.
SEEG · stereoelectroencephalography · causal filtering · missing data · state estimation · ridge regression · brain-computer interfaces
References
6- usesA learned state reduced mean 7 Hz filter phase error after 100 ms simulated gaps in nine SEEG rest records with known initial states sci:sha256:101556a678d7a1131cc51640655ad1744d73bf01ea5d5882f68258510d434efd
We used its 144 original models, selected penalties, feature means, scales, and target means. We rebuilt its unconstrained weights from the early voltage data. New fits kept its model coordinates and penalties.
- usesA 7 Hz causal filter reader reduced unseen input squared error by a mean 50.4 percent at 50 ms in nine SEEG rest records sci:sha256:f851bd1de5c8367288ad9745a4316baf6f4e7fde8dfec5bb56d0adb133df2ac4
We used its causal 7 Hz complex filter and past 27-feature gap reader definition through the parent. The new state constraint applies to that affine reader.
- usesA 0.95 gap map bound limited state sensitivity but increased phase error under dense irregular loss in nine SEEG rest records sci:sha256:3578c11857233541187c42c93f7c8a0fd161d93ce6742fca4a4597bab8f051bc
We used the nine voltage extracts, fixed models, held buffer rule, 0.95 gap map limit, and irregular loss schedules. We reproduced the unchanged dense and fixed baseline scores. We tested new early history and constrained map fits.
- usesTwo exact bounds certified a 7 Hz SEEG filter phase after simulated gaps in nine rest records under voltage caps sci:sha256:1e7436d8e30edf812519ca308906d17bd0b1a757df88c2ccc6c9d70a987f5cb5
We used its public raw prefix reconstruction recipe through the direct parent. Fresh source checks reproduced all nine selected voltage arrays exactly.
- usesdoi:10.18112/openneuro.ds006065.v1.0.0
Seven of the nine analyzed SEEG rest records came from OpenNeuro ds006065 version 1.0.0.
- usesdoi:10.18112/openneuro.ds005670.v1.0.0
Two of the nine analyzed SEEG rest records came from OpenNeuro ds005670 version 1.0.0.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.