Paper detail

Splitting forward-backward penalty scheme for constrained variational problems

We study a forward backward splitting algorithm that solves the variational inequality \begin{equation*} A x +\nabla Φ(x)+ N_C (x) \ni 0 \end{equation*} where $H$ is a real Hilbert space, $A: H\rightrightarrows H$ is a maximal monotone operator, $Φ: H\to\mathbb{R}$ is a smooth convex function, and $N_C$ is the outward normal cone to a closed convex set $C\subset H$. The constraint set $C$ is represented as the intersection of the sets of minima of two convex penalization function $Ψ_1:H\to\mathbb{R}$ and $Ψ_2: H\to\mathbb{R}\cup \{+\infty\}$. The function $Ψ_1$ is smooth, the function $Ψ_2$ is proper and lower semicontinuous. Given a sequence $(β_n)$ of penalization parameters which tends to infinity, and a sequence of positive time steps $(λ_n)$, the algorithm $$ \left\{\begin{array}{rcl} x_1 & \in & H,\\ x_{n+1} & = & (I+λ_n A+λ_nβ_n\partialΨ_2)^{-1}(x_n-λ_n\nablaΦ(x_n)-λ_nβ_n\nablaΨ_1(x_n)),\ n\geq 1. \end{array}\right. $$ performs forward steps on the smooth parts and backward steps on the other parts. Under suitable assumptions, we obtain weak ergodic convergence of the sequence $(x_n)$ to a solution of the variational inequality. Convergence is strong when either $A$ is strongly monotone or $Φ$ is strongly convex. We also obtain weak convergence of the whole sequence $(x_n)$ when $A$ is the subdifferential of a proper lower-semicontinuous convex function. This provides a unified setting for several classical and more recent results, in the line of historical research on continuous and discrete gradient-like systems.