Backward Inertial Modification of the Bregman Golden Ratio Algorithm for Solving Variational Inequalities
摘要
A backward inertial modification of the Bregman golden ratio algorithm is proposed for solving variational inequalities. The algorithm incorporates an inertial step, a Bregman distance, and a fully adaptive stepsize strategy. The convergence of the method is established under standard monotonicity and locally Lipschitz continuity assumptions, and a sublinear convergence rate is derived. Comparative numerical experiments illustrate the efficacy and applicability of the proposed approach, highlighting its advantages over existing techniques.