A note on the stabilizability of stochastic heat equations with multiplicative noiseSur la stabilisabilité des équations de la chaleur stochastiques avec bruit multiplicatif

We show that a stochastic heat equation with multiplicative noise on a bounded domain $\text{D}$ can be stabilized by a control acting only on a subdomain $\text{O}\text{?}\text{D}$ if $\text{D}\text{?}\text{O}$ is sufficiently ‘thin’. We consider both linear and semilinear stochastic heat equations. To cite this article: V. Barbu et al., C. R. Acad. Sci. Paris, Ser. I 334 (2002) 311–316.     

An approximation result for nonlinear SPDEs with Neumann boundary conditions

We establish an approximation result to the solution of a semi linear stochastic partial differential equation with a Neumann boundary condition. Our approach is based on the theory of backward doubly stochastic differential equations. To cite this article: N. Mrhardy, C. R. Acad. Sci. Paris, Ser. I 346 (2008).     

The Neyman–Pearson lemma under g-probability

The Neyman–Pearson fundamental lemma is generalized under g-probability. With convexity assumptions, a sufficient and necessary condition which characterizes the optimal randomized tests is obtained via a maximum principle for stochastic control. To cite this article: S. Ji, X.Y. Zhou, C. R. Acad. Sci. Paris, Ser. I 346 (2008).     

Malliavin calculus for highly degenerate 2D stochastic Navier–Stokes equations

This Note mainly presents the results from “Malliavin calculus and the randomly forced Navier–Stokes equation” by J.C. Mattingly and E. Pardoux. It also contains a result from “Ergodicity of the degenerate stochastic 2D Navier–Stokes equation” by M. Hairer and J.C. Mattingly. We study the Navier–Stokes equation on the two-dimensional torus when forced by a finite dimensional Gaussian white noise. We give conditions under which the law of the solution at any time $t>0$, projected on a finite dimensional subspace, has a smooth density with respect to Lebesgue measure. In particular, our results hold for specific choices of four dimensional Gaussian white noise. Under additional assumptions, we show that the preceding density is everywhere strictly positive. This Note's results are a critical component in the ergodic results discussed in a future article. To cite this article: M. Hairer et al., C. R. Acad. Sci. Paris, Ser. I 339 (2004).     

Density estimates for a random noise propagating through a chain of differential equations

We here provide two sided bounds for the density of the solution of a system of n differential equations of dimension d, the first one being forced by a non-degenerate random noise and the n−1 other ones being degenerate. The system formed by the n equations satisfies a suitable Hörmander condition: the second equation feels the noise plugged into the first equation, the third equation feels the noise transmitted from the first to the second equation and so on … , so that the noise propagates one way through the system. When the coefficients of the system are Lipschitz continuous, we show that the density of the solution satisfies Gaussian bounds with non-diffusive time scales. The proof relies on the interpretation of the density of the solution as the value function of some optimal stochastic control problem.     

Invariant measures for dichotomous stochastic differential equations in Hilbert spaces

We study existence of invariant measures for semilinear stochastic differential equations in Hilbert spaces. We consider infinite dimensional noise that is white in time and colored in space and we assume that the nonlinearities are Lipschitz continuous. We show that if the equation is dichotomous in the sense that the semigroup generated by the linear part is hyperbolic and the Lipschitz constants of the nonlinearities are not too large, then existence of a solution with bounded mean squares implies existence of an invariant measure. To cite this article: O. Van Gaans, S. Verduyn Lunel, C. R. Acad. Sci. Paris, Ser. I 334 (2002) 1083–1088.     

Algorithmes stochastiques à bruit dépendant

We introduce different ways of modeling the dependency of the input noise of stochastic algorithms. We are aimed to prove that such innovations allow us to use the ODE (ordinary differential equation) method. Illustrations in the linear regression framework and in the law of the large number for triangular arrays of weighted dependent random variables are also given. We have aimed to provide results easy to check in practice. To cite this article: P. Doukhan, O. Brandière, C. R. Acad. Sci. Paris, Ser. I 337 (2003).     

Plongement stochastique des systèmes lagrangiens

We define an operator which extends classical differentiation from smooth deterministic functions to certain stochastic processes. Based on this operator, we define a procedure which associates a stochastic analog to standard differential operators and ordinary differential equations. We call this procedure stochastic embedding. By embedding Lagrangian systems, we obtain a stochastic Euler–Lagrange equation which, in the case of natural Lagrangian systems, is called the embedded Newton equation. This equation contains the stochastic Newton equation introduced by Nelson in his dynamical theory of Brownian diffusions. Finally, we consider a diffusion with a gradient drift, a constant diffusion coefficient and having a probability density function. We prove that a necessary condition for this diffusion to solve the embedded Newton equation is that its density be the square of the modulus of a wave function solution of a linear Schrödinger equation. To cite this article: J. Cresson, S. Darses, C. R. Acad. Sci. Paris, Ser. I 342 (2006).     

An exact minimum variance filter for a class of discrete time systems with random parameter perturbations

An exact, closed-form minimum variance filter is designed for a class of discrete time uncertain systems which allows for both multiplicative and additive noise sources. The multiplicative noise model includes a popular class of models (Cox-Ingersoll-Ross type models) in econometrics. The parameters of the system under consideration which describe the state transition are assumed to be subject to stochastic uncertainties. The problem addressed is the design of a filter that minimizes the trace of the estimation error variance. Sensitivity of the new filter to the size of parameter uncertainty, in terms of the variance of parameter perturbations, is also considered. We refer to the new filter as the ‘perturbed Kalman filter’ (PKF) since it reduces to the traditional (or unperturbed) Kalman filter as the size of stochastic perturbation approaches zero. We also consider a related approximate filtering heuristic for univariate time series and we refer to filter based on this heuristic as approximate perturbed Kalman filter (APKF). We test the performance of our new filters on three simulated numerical examples and compare the results with unperturbed Kalman filter that ignores the uncertainty in the transition equation. Through numerical examples, PKF and APKF are shown to outperform the traditional (or unperturbed) Kalman filter in terms of the size of the estimation error when stochastic uncertainties are present, even when the size of stochastic uncertainty is inaccurately identified.     

Stochastic Optimal Control for the Stochastic Heat Equation with Exponentially Growing Coefficients and with Control and Noise on a Subdomain

Abstract

We consider stochastic optimal control problems in Banach spaces, related to nonlinear controlled equations with dissipative non linearities: on the nonlinear term we do not impose any growth condition. The problems are treated via the backward stochastic differential equations approach, that allows also to solve in mild sense Hamilton Jacobi Bellman equations in Banach spaces. We apply the results to controlled stochastic heat equation, in space dimension 1, with control and noise acting on a subdomain.     

Un théorème de représentation des solutions de viscosité d'une équation d'Hamilton–Jacobi–Bellman ergodique dégénérée sur le tore

We consider an ergodic Hamilton–Jacobi–Bellman equation coming from a stochastic control problem in which there are exactly k points where the dynamics vanishes and the Lagrangian is minimal. Under a stabilizability assumption, we state that the solutions of the ergodic equation are uniquely determined by their value on these k points, and that the set of solutions is sup-norm isometric to a non-empty closed convex set whose dimension is less or equal to k. To cite this article: M. Akian et al., C. R. Acad. Sci. Paris, Ser. I 346 (2008).     

Une classification des flots solutions d'une EDS

Under general hypotheses, we show that the flows of kernels can be associated to a stochastic differential equation (SDE). We also show a classification theorem of the solutions of the SDE: they can be obtained through filtering the coalescing solution with respect to a sub-noise containing the white noise driving the SDE. The example of the isotropic flows is studied. To cite this article: Y. Le Jan, O. Raimond, C. R. Acad. Sci. Paris, Ser. I 336 (2003).     

Inversion of nonlinear stochastic operators

The operator-theoretic method (Adomian and Malakian, J. Math. Anal. Appl.76(1), (1980), 183–201) recently extended Adomian's solutions of nonlinear stochastic differential equations (G. Adomian, Stochastic Systems Analysis, in “Applied Stochastic Processes,” Nonlinear Stochastic Differential Equations, J. Math. Anal. Appl.55(1) (1976), 441–452; On the modeling and analysis of nonlinear stochastic systems, in “Proceeding, International Conf. on Mathematical Modeling.” Vol. 1, pp. 29–40) to provide an efficient computational procedure for differential equations containing polynomial, exponential, and trigonometric nonlinear terms N(y). The procedure depends on the calculation of certain quantities A_n and B_n. This paper generalizes the calculation of the A_n and B_n to much wider classes of nonlinearities of the form N(y, y′,…). Essentially, the method provides a systematic computational procedure for differential equations containing any nonlinear terms of physical significance. This procedure depends on a recurrence rule from which explicit general formulae are obtained for the quantities A_n and B_n for any order n in a convenient form. This paper also demonstrates the significance of the iterative series decomposition proposed by Adomian for linear stochastic operators in 1964 and developed since 1976 for nonlinear stochastic operators. Since both the nonlinear and stochastic behavior is quite general, the results are extremely significant for applications. Processes need not, for example, be limited to either Gaussian processes, white noise, or small fluctuations.     

A class of Gaussian processes with fractional spectral measures

We study a family of stationary increment Gaussian processes, indexed by time. These processes are determined by certain measures σ (generalized spectral measures), and our focus here is on the case when the measure σ is a singular measure. We characterize the processes arising from σ when σ is in one of the classes of affine selfsimilar measures. Our analysis makes use of Kondratiev white noise spaces. With the use of a priori estimates and the Wick calculus, we extend and sharpen (see Theorem 7.1) earlier computations of Ito stochastic integration developed for the special case of stationary increment processes having absolutely continuous measures. We further obtain an associated Ito formula (see Theorem 8.1).     

Asymptotic estimates for white noise distributions

Let θ be a Young function. Using properties of the Laplace and Legendre transforms, it is shown that white noise measures in the dual of a test function space of θ-exponential growth satisfy an exponential decay property with rate θ. An application to stochastic differential equations is given. To cite this article: H. Ouerdiane, N. Privault, C. R. Acad. Sci. Paris, Ser. I 338 (2004).     

On pth roots of stochastic matrices

In Markov chain models in finance and healthcare a transition matrix over a certain time interval is needed but only a transition matrix over a longer time interval may be available. The problem arises of determining a stochastic pth root of a stochastic matrix (the given transition matrix). By exploiting the theory of functions of matrices, we develop results on the existence and characterization of matrix pth roots, and in particular on the existence of stochastic pth roots of stochastic matrices. Our contributions include characterization of when a real matrix has a real pth root, a classification of pth roots of a possibly singular matrix, a sufficient condition for a pth root of a stochastic matrix to have unit row sums, and the identification of two classes of stochastic matrices that have stochastic pth roots for all p. We also delineate a wide variety of possible configurations as regards existence, nature (primary or nonprimary), and number of stochastic roots, and develop a necessary condition for existence of a stochastic root in terms of the spectrum of the given matrix.     

Flots de noyaux et flots coalescents

We present a part of the results of Le Jan and Raimond (math.PR/9909147). We show that starting with a compatible family of Feller semigroups, one can construct a stochastic flow of kernels. Under an additional hypotheses (on the 2-points motion), we show that it is possible to associate to a flow of kernels a coalescing flow such that the flow of kernels can be obtained by filtering the coalescing flow with respect to a sub-noise of an extension of the noise generated by the coalescing flow. To cite this article: Y. Le Jan, O. Raimond, C. R. Acad. Sci. Paris, Ser. I 336 (2003).     

Natural gradient algorithm for stochastic distribution systems with output feedback

In this paper, we use natural gradient algorithm to control the shape of the conditional output probability density function for the stochastic distribution systems from the viewpoint of information geometry. The considered system here is of multi-input and single output with an output feedback and a stochastic noise. Based on the assumption that the probability density function of the stochastic noise is known, we obtain the conditional output probability density function whose shape is only determined by the control input vector under the condition that the output feedback is known at any sample time. The set of all the conditional output probability density functions forms a statistical manifold (M), and the control input vector and the output feedback are considered as the coordinate system. The Kullback divergence acts as the distance between the conditional output probability density function and the target probability density function. Thus, an iterative formula for the control input vector is proposed in the sense of information geometry. Meanwhile, we consider the convergence of the presented algorithm. At last, an illustrative example is utilized to demonstrate the effectiveness of the algorithm.     

Jeux à champ moyen. II – Horizon fini et contrôle optimal

We continue in this Note our study of the notion of mean field games that we introduced in a previous Note. We consider here the case of Nash equilibria for stochastic control type problems in finite horizon. We present general existence and uniqueness results for the partial differential equations systems that we introduce. We also give a possible interpretation of these systems in term of optimal control. To cite this article: J.-M. Lasry, P.-L. Lions, C. R. Acad. Sci. Paris, Ser. I 343 (2006).