LATIDIA · Ciberseguridad

SciCheck: un marco para verificar la integridad de la capacitación en aprendizaje profundo

arXiv: 2611.09889v1Tipo de anuncio: nuevo Resumen: La capacitación en aprendizaje profundo es difícil de verificar dentro de un flujo de trabajo científico: su arquitectura, hiperparámetros y procedimiento rara vez se fijan hasta que la capacitación se ha realizado.

La noticia

arXiv:2611.09889v1 Announce Type: new Abstract: Deep learning training is difficult to verify within a scientific workflow: its architecture, hyperparameters, and procedure are rarely fixed until training has been attempted, and two honest runs never produce the same weights, so recomputation cannot serve as a check. We present SciCheck, a framework in which the scientist explores training configurations freely and posts a blueprint to a provenance ledger for each candidate they want verified, and a verifying scientist judges the resulting run in two layers: a statistical judge model that compares per-step gradient statistics against the verifier's own reference run, and a spot-check that re-executes randomly chosen training segments whose checkpoints are