Tight Convergence Rate Bounds for Optimization Under Power Law Spectral Conditions.
Maksim VelikanovDmitry YarotskyPublished in: CoRR (2022)
Keyphrases
- convergence rate
- power law
- global convergence
- faster convergence rate
- upper bound
- lower bound
- number of iterations required
- convergence speed
- step size
- learning rate
- worst case
- scale free
- gradient method
- small world
- mutation operator
- primal dual
- optimization algorithm
- global search
- numerical stability
- power law distribution
- sufficient conditions
- global optimization
- wavelet neural network
- combinatorial optimization
- lp norm
- optimization problems
- recommender systems
- learning algorithm