Predictable history-adaptive virtual perturbations yield information-theoretic generalization bounds for SGD that incorporate adaptive noise geometries without altering the algorithm.
Lecture 6.5—RMSProp: Divide the gradient by a running average of its recent magnitude
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Information-Theoretic Generalization Bounds for Stochastic Gradient Descent with Predictable Virtual Noise
Predictable history-adaptive virtual perturbations yield information-theoretic generalization bounds for SGD that incorporate adaptive noise geometries without altering the algorithm.