NeurIPS 2025 Explorer
Concepts
Authors
Glossary
johnsanterre.github.io
sub-optimality gap
The difference between the performance of a model or solution and the best possible performance, providing insight into the efficiency and potential improvements of an algorithm.
4 papers
Does Stochastic Gradient really succeed for bandits?
Does Stochastic Gradient really succeed for bandits?
Individual Regret in Cooperative Stochastic Multi-Armed Bandits
Pareto Optimal Risk-Agnostic Distributional Bandits with Heavy-Tail Rewards