Algorithms with logarithmic or sublinear regret for constrained contextual bandits

Keyphrases

• Sub-linear Regret 100%
• Contextual Bandits 100%
• Logarithmic Regret 100%
• Expected Payoff 60%
• Time Constraints 40%
• Budget Constraint 40%
• Linear Programming 40%
• Upper Confidence Bound 40%
• Context Distribution 40%
• Computationally Efficient Algorithms 20%
• Algorithm Design 20%
• Distribution Cost 20%
• Boundary Cases 20%
• Oracle 20%
• Costing System 20%
• General Systems 20%
• Unit Cost 20%
• Binding Method 20%
• Near-optimality 20%
• Linear Algorithm 20%
• Complex Coupling 20%
• Regret Bounds 20%
• Heterogeneous Costs 20%

Computer Science

• Linear Programming 100%
• Time Constraint 66%
• Budget Constraint 66%
• Confidence Bound 66%
• Approximation (Algorithm) 33%
• Efficient Algorithm 33%
• Algorithm Design 33%