arXiv subjects
Publications and source records attributed to Andrzej Święch.
We prove the Aleksandrov-Bakelman-Pucci maximum principle for $L^n$-viscosity sub/super-solutions of fully nonlinear uniformly elliptic equations in $Ω\subset\mathbb R^n$ with the drift coefficient $γ$ such that $γ(\ln(1+γ))^{\frac{n-1}{n}}\in L^n(Ω)$. The approach is based on convex duality and Trudinger-Moser type inequalities.
Optimal control of interacting particles governed by stochastic evolution equations in Hilbert spaces is an open area of research. Such systems naturally arise in formulations where each particle is modeled by stochastic partial differential equations, path-dependent stochastic differential equations (such as stochastic delay differential equations or stochastic Volterra integral equations), or partially observed stochastic systems. The purpose of this manuscript is to build a limiting theory as the number of particles tends to infinity via the dynamic programming approach. We prove the convergence of the value functions $u_n$ of finite particle systems to a function $\mathcal{V}$, which is the unique $L$-viscosity solution of the corresponding mean-field Hamilton-Jacobi-Bellman equation in the space of probability measures, and we identify its lift with the value function $U$ of the so-called ``lifted'' limit optimal control problem. Under suitable additional assumptions, we show $C^{1,1}$-regularity of $U$, we prove that $\mathcal{V}$ projects precisely onto the value functions $u_n$, and that optimal (resp. optimal feedback) controls of the particle system correspond to optimal (resp. optimal feedback) controls of the lifted control problem started at the corresponding initial condition. To the best of our knowledge, these are the first results of this kind for stochastic optimal control problems for interacting particle systems of stochastic evolution equations in Hilbert spaces. We apply the developed theory to problems arising in economics where the particles are modeled by stochastic delay differential equations and stochastic partial differential equations.
We study a degenerate second order mean field game (MFG) system in a Hilbert space $H$ which couples a Fokker--Planck equation describing the evolution of probability measures on $H$ with a Hamilton--Jacobi--Bellman (HJB) equation for the value function. Our main result establishes existence and uniqueness of solutions to this coupled system. Solutions of the HJB equation are interpreted in the viscosity sense. For existence, we extend the classical fixed-point approach based on Tikhonov's theorem to our setting. A central difficulty in this approach is proving uniqueness for the corresponding linear degenerate Fokker--Planck equation. To address this issue, we introduce a class of suitable adjoint equations and employ viscosity solution techniques to construct sufficiently regular solutions. Uniqueness for the full MFG system is then obtained via an adaptation of the Lasry--Lions monotonicity method.
We study a Mean Field Games (MFG) system in a real, separable infinite dimensional Hilbert space. The system consists of a second order parabolic type equation, called Hamilton-Jacobi-Bellman (HJB) equation in the paper, coupled with a nonlinear Fokker-Planck (FP) equation. Both equations contain a Kolmogorov operator. Solutions to the HJB equation are interpreted in the mild solution sense and solutions to the FP equation are interpreted in an appropriate weak sense. We prove well-posedness of the considered MFG system under certain conditions. The existence of a solution to the MFG system is proved using Tikhonov's fixed point theorem in a proper space. Uniqueness of solutions is obtained under typical separability and Lasry-Lions type monotonicity conditions.
This paper continues the study of controlled interacting particle systems with common noise started in [W. Gangbo, S. Mayorga and A. Święch, SIAM J. Math. Anal. 53 (2021), no. 2, 1320--1356] and [S. Mayorga and A. Święch, SIAM J. Control Optim. 61 (2023), no. 2, 820--851]. First, we extend the following results of the previously mentioned works to the case of multiplicative noise: (i) We generalize the convergence of the value functions $u_n$ corresponding to control problems of $n$ particles to the value function $V$ corresponding to an appropriately defined infinite dimensional control problem; (ii) we prove, under certain additional assumptions, $C^{1,1}$ regularity of $V$ in the spatial variable. The second main contribution of the present work is the proof that if $DV$ is continuous (which, in particular, includes the previously proven case of $C^{1,1}$ regularity in the spatial variable), the value function $V$ projects precisely onto the value functions $u_n$. Using this projection property, we show that optimal controls of the finite dimensional problem correspond to optimal controls of the infinite dimensional problem and vice versa. In the case of a linear state equation, we are able to prove that $V$ projects precisely onto the value functions $u_n$ under relaxed assumptions on the coefficients of the cost functional by using approximation techniques in the Wasserstein space, thus covering cases where $V$ may not be differentiable.
We study optimal control problems governed by abstract infinite dimensional stochastic differential equations using the dynamic programming approach. In the first part, we prove Lipschitz continuity, semiconcavity and semiconvexity of the value function under several sets of assumptions, and thus derive its $C^{1,1}$ regularity in the space variable. Based on this regularity result, we construct optimal feedback controls using the notion of the $B$-continuous viscosity solutions for the associated Hamilton--Jacobi--Bellman equation. This is done in the case when the noise coefficient is independent of the control variable. We also discuss applications of our results to optimal control problems governed by stochastic reaction-diffusion equations and, under economic motivations, stochastic delay differential equations.
In this manuscript, we study optimal control problems for stochastic delay differential equations using the dynamic programming approach in Hilbert spaces via viscosity solutions of the associated Hamilton-Jacobi-Bellman equations. We show how to use the partial $C^{1,α}$-regularity of the value function established in \cite{defeo_federico_swiech} to obtain optimal feedback controls. The main result of the paper is a verification theorem which provides a sufficient condition for optimality using the value function. We then discuss its applicability to the construction of optimal feedback controls. We provide an application to stochastic optimal advertising problems.
In this manuscript we consider a class optimal control problem for stochastic differential delay equations. First, we rewrite the problem in a suitable infinite-dimensional Hilbert space. Then, using the dynamic programming approach, we characterize the value function of the problem as the unique viscosity solution of the associated infinite-dimensional Hamilton-Jacobi-Bellman equation. Finally, we prove a $C^{1,α}$-partial regularity of the value function. We apply these results to path dependent financial and economic problems (Merton-like portfolio problem and optimal advertising).
We study a free transmission problem driven by degenerate fully nonlinear operators. Our first result concerns the existence of solutions to the associated Dirichlet problem. By framing the equation in the context of viscosity inequalities, we produce optimal regularity results for viscosity solutions and certain strong solutions to the problem. Our findings include regularity in $ C^{1,α}$ spaces, and an explicit characterization of $α$ in terms of the degeneracy rates. We argue by perturbation methods, relating our problem to a homogeneous, fully nonlinear uniformly elliptic equation.
We study an equation governed by a discontinuous fully nonlinear operator. Such discontinuities are solution-dependent, which introduces a free boundary. Working under natural assumptions, we prove the existence of $L^p$-viscosity and strong solutions to the problem. The operator does not satisfy the usual structure conditions and to obtain the existence of solutions we resort to solving approximate problems, combined with an iterative scheme. We believe our strategy can be applied to other classes of non-variational free boundary problems.
We consider an initial value problem for a Hamilton--Jacobi equation with a quadratic and degenerate Hamiltonian. Our Hamiltonian comes from the dynamics of $N$-peakon in the Camassa--Holm equation. It is given by a quadratic form with a singular positive semi-definite matrix. Such a problem does not fall into the standard theory of viscosity solutions. Also viability related results, sometimes used to deal with degenerate Hamiltonians, do not seem applicable in our case. We prove the global existence of a viscosity solution by looking at the associated optimal control problem and showing that the value function is a viscosity solution. The most complicated part is the continuity of a viscosity solution which is obtained in the two-peakon case only. The source of the difficulties is the non-uniqueness of solutions to the state equation in the optimal control problem. We prove that the viscosity solution is Lipschitz continuous and unique on some short time interval if the initial condition is Lipschitz continuous. We end the paper with an example showing the loss of Lipschitz continuity of a viscosity solution in the one-dimensional case.