A primal-dual framework with adaptive dual regularizer achieves O(√T) regret and O(√T log T) constraint violation for constrained OCO without Slater's condition under stochastic constraints, with extensions to adversarial constraints and strongly convex losses.
An optimistic algo- rithm for online convex optimization with adversarial constraints,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Introduces first algorithm for interval regret scaling with gradient variation via two-layer ensemble, plus Lipschitz-smoothness agnostic variant, with extensions to dynamic regret and stochastic settings.
citing papers explorer
-
Constrained Online Convex Optimization without Slater's Condition
A primal-dual framework with adaptive dual regularizer achieves O(√T) regret and O(√T log T) constraint violation for constrained OCO without Slater's condition under stochastic constraints, with extensions to adversarial constraints and strongly convex losses.
-
Online Learning with Gradient-Variation Interval Regret
Introduces first algorithm for interval regret scaling with gradient variation via two-layer ensemble, plus Lipschitz-smoothness agnostic variant, with extensions to dynamic regret and stochastic settings.