regret minimization
A framework in decision-making and game theory aimed at minimizing the difference between the chosen strategy's performance and the best possible performance in hindsight. It is essential in developing algorithms that learn from past decisions.
- $O(\sqrt{T})$ Static Regret and Instance Dependent Constraint Violation for Constrained Online Convex Optimization
- An Improved Algorithm for Adversarial Linear Contextual Bandits via Reduction
- Beyond $\tilde{O}(\sqrt{T})$ Constraint Violation for Online Convex Optimization with Adversarial Constraints
- Design-Based Bandits Under Network Interference: Trade-Off Between Regret and Statistical Inference
- From Contextual Combinatorial Semi-Bandits to Bandit List Classification: Improved Sample Complexity with Sparse Rewards
- Gaussian Process Upper Confidence Bound Achieves Nearly-Optimal Regret in Noise-Free Gaussian Process Bandits
- Improved Regret Bounds for Gaussian Process Upper Confidence Bound in Bayesian Optimization
- Improved Regret Bounds for Gaussian Process Upper Confidence Bound in Bayesian Optimization
- Improved Regret and Contextual Linear Extension for Pandora's Box and Prophet Inequality
- No-Regret Online Autobidding Algorithms in First-price Auctions
- Non-stationary Bandit Convex Optimization: A Comprehensive Study
- Online Bilateral Trade With Minimal Feedback: Don’t Waste Seller’s Time
- Principled Model Routing for Unknown Mixtures of Source Domains
- Regret Bounds for Adversarial Contextual Bandits with General Function Approximation and Delayed Feedback
- Stochastic Regret Guarantees for Online Zeroth- and First-Order Bilevel Optimization
- Tackling Biased Evaluators in Dueling Bandits
- Thompson Sampling for Multi-Objective Linear Contextual Bandit