Dynamical systems method of gradient type for solving nonlinear equations with monotone operators (Q616166)

A version of the Dynamical Systems Method (DSM) of gradient type is studied for solving the equation (1) \(F(u)= f\), where \(F\) is a nonlinear Fréchet differentiable, monotone operator in Hilbert space. If \(F(u)\) is not boundedly invertible then solving (1) for a given noisy \(f_\delta\) may be a ill-posed problem. Without extra assumptions, usually source-type assumptions about the right-hand side or some assumption concerning the smoothness of the solution, one cannot get a specific rate of convergence. Since such assumptions are difficult to verify and often they do not hold, then a Discrepancy Principle (DP) is proposed here. On the basis DP and DSM the convergence of an iterative scheme is formulated and the convergence of this scheme to the minimal-norm solution is proved under weaker assumption on the differentiability of \(F\) and for a larger class of regularizing functions. It is demonstrated that the proposed iterative scheme yields good numerical results.