Gradient-Variation Online Adaptivity for Accelerated Optimization with Hölder Smoothness
accelerated convergenceaccelerationadaptivityconvex optimizationdetection-based proceduregradient-variation algorithmgradient-variation regretguess-and-checkhölder functionslipschitz functionsonline learningonline-to-batch conversionsmoothnessstochastic convex optimizationuniversal method
Smoothness is known to be crucial for acceleration in offline optimization, and for gradient-variation regret minimization in online learning. Interestingly, these two problems are actually closely connected