18 May 2018 in Optimization Methods and Software
In this paper, we propose a generic algorithmic framework for stochastic proximal quasi-Newton (SPQN) methods to solve non-convex composite optimization problems. Stochastic second-order information is explored to construct proximal subproblem. Under mild conditions we show the non-asympotic convergence of the proposed algorithm to stationary point of original problems and analyse its computational complexity. Besides, we extend the proximal form of Polyak-Łojasiewicz (PL) inequality to constrained settings and obtain the constrained proximal PL (CP-PL) inequality. Under CP-PL inequality linear convergence rate of the proposed algorithm is achieved. Moreover, we propose a modified self-scaling symmetric rank one incorporated in the framework for SPQN method, which is called stochastic symmetric rank one method. Finally, we report some numerical experiments to reveal the effectiveness of the proposed algorithm.
15 July 2014 in Optimization Methods and Software
In this paper we propose an augmented Lagrangian trust region method for equality constrained optimization. Different from standard augmented Lagrangian methods which minimize the augmented Lagrangian function for fixed Lagrange multiplier and penalty parameter at each iteration, the proposed method tries to minimize its second-order approximation function. We propose a new strategy for adjusting the penalty parameter. With adaptive update of Lagrange multipliers, we prove the global convergence of the proposed method. Numerical results on test problems from the CUTEr collection are also reported.
24 June 2022 in Computational Optimization and Applications
1 January 2026 in Mathematics of Operations Research
Nonconvex constrained stochastic optimization has emerged in many important application areas. Subject to general functional constraints, it minimizes the sum of an expectation function and a nonsmooth regularizer. Main challenges arise because of the stochasticity in the random integrand and the possibly nonconvex functional constraints. To address these issues, we propose a momentum-based linearized augmented Lagrangian method (MLALM). MLALM adopts a single-loop framework and incorporates a recursive momentum scheme to compute the stochastic gradient, which enables the construction of a stochastic approximation to the augmented Lagrangian function. We provide an analysis of global convergence of MLALM. Under mild conditions and with unbounded penalty parameters, we show that the sequences of average stationarity measure and constraint violations are convergent in expectation. Under a constraint qualification assumption, the sequences of average constraint violation and complementary slackness measure converge to zero in expectation. We also explore properties of those related metrics when penalty parameters are bounded. Furthermore, we investigate oracle complexities of MLALM in terms of the total number of stochastic gradient evaluations to find an ϵ-stationary point and an ϵ-Karush -Kuhn -Tucker point when assuming the constraint qualification. Numerical experiments on two types of test problems reveal promising performances of the proposed algorithm. Funding: This work was supported by the National Natural Science Foundation of China [Grant 12271278], the Major Key Project of PCL [Grant PCL2022A05], and the Natural Science Foundation of Shanghai [Grant 21ZR1442800].
11 August 2017 in Computational Optimization and Applications
17 February 2020 in Optimization Letters