Convergence rates of the stochastic alternating algorithm for bi-objective optimization
From MaRDI portal
Abstract: Stochastic alternating algorithms for bi-objective optimization are considered when optimizing two conflicting functions for which optimization steps have to be applied separately for each function. Such algorithms consist of applying a certain number of steps of gradient or subgradient descent on each single objective at each iteration. In this paper, we show that stochastic alternating algorithms achieve a sublinear convergence rate of , under strong convexity, for the determination of a minimizer of a weighted-sum of the two functions, parameterized by the number of steps applied on each of them. An extension to the convex case is presented for which the rate weakens to . These rates are valid also in the non-smooth case. Importantly, by varying the proportion of steps applied to each function, one can determine an approximation to the Pareto front.
Recommendations
- Optimization over the Pareto outcome set associated with a convex bi-objective optimization problem: theoretical results, deterministic algorithm and application to the stochastic case
- The stochastic multi-gradient algorithm for multi-objective optimization and its application to supervised machine learning
- Rates of convergence of semi-stochastic approximation procedures for solving stochastic optimization problems
- On efficiency of a single variable bi-objective optimization algorithm
- A stochastic multiple gradient descent algorithm
Cites work
- A projected gradient method for vector optimization problems
- A proximal point-type method for multicriteria optimization
- A quadratically convergent Newton method for vector optimization
- A steepest descent method for vector optimization
- A stochastic multiple gradient descent algorithm
- A subgradient method for multiobjective optimization on Riemannian manifolds
- A subgradient method for vector optimization problems
- A trust-region method for unconstrained multiobjective problems with applications in satisficing processes
- Direct Multisearch for Multiobjective Optimization
- Fair coresets and streaming algorithms for fair \(k\)-means
- First-order methods in optimization
- Inexact projected gradient method for vector optimization
- Multiple-gradient descent algorithm (MGDA) for multiobjective optimization
- Multiple-gradient descent algorithm for Pareto-front identification
- Newton's method for multiobjective optimization
- Nonlinear Conjugate Gradient Methods for Vector Optimization
- On a Bicriterion Formulation of the Problems of Integrated System Identification and System Optimization
- Optimization methods for large-scale machine learning
- Proximal gradient methods for multiobjective optimization and their applications
- Proximal Methods in Vector Optimization
- Quasi-Newton methods for solving multiobjective optimization
- Quasi-Newton's method for multiobjective optimization
- Recent advances in evolutionary multi-objective optimization
- Robust Stochastic Approximation Approach to Stochastic Programming
- Solution approaches for the multiobjective stochastic programming
- Steepest descent methods for multicriteria optimization.
- Stochastic approach versus multiobjective approach for obtaining efficient solutions in stochastic multiobjective programming problems
- Trust region methods for solving multiobjective optimisation
Cited in
(2)
This page was built for publication: Convergence rates of the stochastic alternating algorithm for bi-objective optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6108978)
