Dynamic Regret Bounds for Constrained Online Nonconvex Optimization Based on Polyak-Lojasiewicz Regions

Keyphrases

• Optimization Approach 100%
• Dynamic Regret 100%
• Regret Bounds 100%
• Online Non-convex Optimization 100%
• Global Solution 75%
• Optimal Decision 50%
• Loss Function 50%
• Decision Maker 25%
• Nonconvex 25%
• Numerical Experiments 25%
• Temporal Variation 25%
• Algorithmic Performance 25%
• Nonconvex Optimization Problems 25%
• Gradient Projection 25%
• Online Optimization Problems 25%
• Ojasiewicz Inequality 25%

Mathematics

• Global Solution 100%
• Optimal Decision 66%
• Loss Function 66%
• Objective Function 33%
• Decision Maker 33%
• Numerical Experiment 33%
• Desirable Property 33%
• Convex Case 33%

Engineering

• Regret Dynamic 100%
• Optimal Decision 50%
• Optimisation Problem 50%
• Loss Function 50%
• Objective Function 25%
• Decision Maker 25%
• Numerical Experiment 25%
• Temporal Variation 25%
• Gradient Descent 25%