Gradient Primal-Dual Algorithm Converges to Second-Order Stationary Solution for Nonconvex Distributed Optimization Over Networks.
Mingyi HongMeisam RazaviyaynJason D. LeePublished in: ICML (2018)
Keyphrases
- primal dual
- convex optimization
- linear programming
- nonlinear programming
- variational inequalities
- linear programming problems
- linear program
- convex programming
- saddle point
- interior point algorithm
- optimization problems
- affine scaling
- interior point methods
- dual formulation
- quadratic program
- simplex method
- optimization method
- line search
- convex optimization problems
- convergence rate
- optimal solution
- approximation algorithms
- semidefinite programming
- augmented lagrangian method
- algorithm for linear programming
- interior point
- convex relaxation
- optimality conditions
- convex functions
- quadratic programming
- simplex algorithm
- objective function
- evolutionary algorithm
- inequality constraints
- mathematical program
- sensitivity analysis
- multiscale
- semidefinite
- duality gap
- high resolution
- special case
- knapsack problem
- augmented lagrangian
- mixed integer
- image segmentation