Unknown distance to optimality does not increase the sample complexity of stochastic convex optimization; a two-stage regularized ERM plus constrained optimization matches optimal known-parameter bounds.
The ladder: A reliable leaderboard for machine learn- ing competitions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
The Sample Complexity of Parameter-Free Stochastic Convex Optimization
Unknown distance to optimality does not increase the sample complexity of stochastic convex optimization; a two-stage regularized ERM plus constrained optimization matches optimal known-parameter bounds.