Stabilizing updates in differentially private stochastic gradient descent with buffered rejection
Artículo
Material complementario disponible
Artículo
DOAJ
Abstract Differentially private stochastic gradient descent is a standard algorithm for training deep models on sensitive data, but under tight privacy budgets it must add large noise to every step, which slows convergen...
LCC LCC:Medicine; TENDOlNjaWVuY2U~Idioma eng
Material complementario disponibleEl enlace apunta a material asociado, anexos, tablas, datos o página complementaria. No se marca como libro/texto completo.
Material complementario