Under a model-loss decomposition, clipping model outputs in tail events yields adaptive SCO bounds matching known-parameter optima up to logarithmic factors in uncertainty.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
Develops constant-stepsize and auto-conditioned projected gradient methods plus stochastic variants that achieve new iteration complexity bounds for finding approximate stationary points in nonconvex smooth optimization.
HDET lets data-parallel replicas explore a spread of learning rates independently before averaging parameters, with an auto-LR controller driven by inter-replica loss differences to produce a self-adapting schedule without extra sweeps.
citing papers explorer
-
Clipping the Price of Adaptivity at the Tail
Under a model-loss decomposition, clipping model outputs in tail events yields adaptive SCO bounds matching known-parameter optima up to logarithmic factors in uncertainty.
-
Projected gradient methods for nonconvex and stochastic smooth optimization: new complexities and auto-conditioned stepsizes
Develops constant-stepsize and auto-conditioned projected gradient methods plus stochastic variants that achieve new iteration complexity bounds for finding approximate stationary points in nonconvex smooth optimization.
-
Scalable Hyperparameter-Divergent Ensemble Training with Automatic Learning Rate Exploration for Large Models
HDET lets data-parallel replicas explore a spread of learning rates independently before averaging parameters, with an auto-LR controller driven by inter-replica loss differences to produce a self-adapting schedule without extra sweeps.