On the discrepancy principle for the dynamical systems method
A. G. Ramm
Abstract
Assume that Au=f, (1) is a solvable linear equation in a Hilbert space, ||A||<∞, and R(A) is not closed, so problem (1) is ill-posed. Here R(A) is the range of the linear operator A. A DSM (dynamical systems method) for solving (1), consists of solving the following Cauchy problem: u= -u +(B+(t))-1A*f, u(0)=u0, (2) where B:=A*A, u:= dudt, u0 is arbitrary, and (t)>0 is a continuously differentiable function, monotonically decaying to zero as t ∞. A.G.Ramm has proved that, for any u0, problem (2) has a unique solution for all t>0, there exists y:=w(∞):=t ∞u(t), Ay=f, and y is the unique minimal-norm solution to (1). If f is given, such that ||f-f||≤ , then u(t) is defined as the solution to (2) with f replaced by f. The stopping time is defined as a number t that 0||u (t)-y||=0, and 0t=∞. A discrepancy principle is proposed and proved in this paper. This principle yields t as the unique solution to the equation: ||A(B+(t))-1A*f -f||=, (3) where it is assumed that ||f||> and f N(A*). For nonlinear monotone A a discrepancy principle is formulated and justified.
Create a lesson
Related papers
Physical and emergent nonpairwise interactions in oscillator networks: from higher-order phase reduction to coupling design
Riccardo Muolo, Hiroya Nakao, Christian Bick
Degree Growth of Iterates of Curves and Likely Intersections
Sina Saleh, Jit Wu Yap
Infinite prime sumsets in structured and Uk(Φ)-uniform sets
Felipe Hernández, Tristán Radić
Long-Lived Carpet-Like Transients in Time-Dependent Modular Discrete Laplacian Dynamics
Małgorzata Nowak-Kępczyk
SIPHy: Sparse identification of port-Hamiltonian systems from noisy data
Håkon Noren Myhr, Sølve Eidnes, J. Nathan Kutz
Extended dynamic mode decomposition with Fourier dictionaries: Error bounds and fast implementation
Felix Bartel, Sandra Ritter, Manuel Schaller et al.