N. V. Krylov

Elliptic equations in Sobolev spaces with Morrey drift and the zeroth-order coefficients

We consider elliptic equations with operators $L=a^{ij}D_{ij}+b^{i}D_{i}-c$ with $a$ being almost in VMO, $b$ in a Morrey class containing $L_{d}$ , and $c\geq0$ in a Morrey class containing $L_{d/2}$ . We prove the solvability in Sobolev spaces of $Lu=f\in L_{p}$ in bounded $C^{1,1}$ -domains, and of $λu-Lu=f$ in the whole space for any $λ>0$ . Weak uniqueness of the martingale problem associated with such operators is also discussed.

Estimates in $L_{p}$ for solutions of SPDEs with coefficients in Morrey classes

For solutions of a certain class of SPDEs in divergence form we present some estimates of their $L_{p}$ -norms and the $L_{p}$ -norms of their first-order derivatives. The main novelty is that the low-order coefficients are supposed to belong to certain Morrey classes instead of $L_{p}$ -spaces. Our results are new even if there are no stochastic terms in the equation.

Linear and fully nonlinear elliptic equations with Morrey drift

We present some results concerning the solvability of linear elliptic equations in bounded domains with the main coefficients almost in VMO, the drift and the free terms in Morrey classes containing $L_{d}$ , and bounded zeroth order coefficient. We prove that the second-order derivatives of solutions are in a local Morrey class containing $W^{2}_{p,loc}$ . Actually, the exposition is given for fully nonlinear equations and encompasses the above mentioned results, which are new even if the main part of the equation is just the Laplacian.

On parabolic Adams's theorem and some other results related to parabolic Morrey spaces

We present several results of embedding type for parabolic Morrey spaces with or without mixed norms. Some other interpolation results for parabolic Morrey spaces are also given.

On strong solutions of Itô's equations with $Dσ$ and $b$ in Morrey classes containing $L_{d}$

We consider Itô uniformly nondegenerate equations with time independent coefficients, the diffusion coefficient in $W^{1}_{2+\varepsilon,loc}$ , and the drift in a Morrey class containing $L_{d}$ . We prove the unique strong solvability in the class of admissible solutions for any starting point. The result is new even if the diffusion is constant.

preprint2021arXiv

On diffusion processes with drift in $L_{d+1}$

This paper is a natural continuation of \cite{Kr_20_2} and \cite{Kr_21_1} where strong Markov processes are constructed in time inhomogeneous setting with Borel measurable uniformly bounded and uniformly nondegenerate diffusion and drift in $L_{d+1}(\mathbb{R}^{d+1})$ and some properties of their Green's functions and probability of passing through narrow tubes are investigated. On the basis of this here we study some further properties of these processes such as Harnack inequality, Hölder continuity of potentials, Fanghua Lin estimates and so on.

preprint2021arXiv

On potentials of Itô's processes with drift in $L_{d+1}$

This paper is a natural continuation of \cite{Kr_20_2}, where strong Markov processes are constructed in time inhomogeneous setting with Borel measurable uniformly bounded and uniformly nondegenerate diffusion and drift in $L_{d+1}(\mathbb{R}^{d+1})$ . Here we study some properties of these processes such as the probability to pass through narrow tubes, higher summability of Green's functions, and so on. The results seem to be new even if the diffusion is constant.

preprint2021arXiv

On the heat equation with drift in $L_{d+1}$

In this paper we deal with the heat equation with drift in $L_{d+1}$ . Basically, we prove that, if the free term is in $L_{q}$ with high enough $q$ , then the equation is uniquely solvable in a rather unusual class of functions such that $\partial_{t}u, D^{2}u\in L_{p}$ with $p<d+1$ and $Du\in L_{q}$ .

Elliptic equations with VMO a, b $\,\in L_{d}$ , and c $\,\in L_{d/2}$

We consider elliptic equations with operators $L=a^{ij}D_{ij}+b^{i}D_{i}-c$ with $a$ being almost in VMO, $b\in L_{d}$ and $c\in L_{q}$ , $c\geq0$ , $d>q\geq d/2$ . We prove the solvability of $Lu=f\in L_{p}$ in bounded $C^{1,1}$ -domains, $1<p\leq q$ , and of $λu-Lu=f$ in the whole space for any $λ>0$ . Weak uniqueness of the martingale problem associated with such operators is also obtained.

Linear and fully nonlinear elliptic equations with $L_{d}$ -drift

In subdomains of $\mathbb{R}^{d}$ we consider uniformly elliptic equations $H\big(v( x),D v( x),D^{2}v( x), x\big)=0$ with the growth of $H$ with respect to $|Dv|$ controlled by the product of a function from $L_{d}$ times $|Dv|$ . The dependence of $H$ on $x$ is assumed to be of BMO type. Among other things we prove that there exists $d_{0}\in(d/2,d)$ such that for any $p\in(d_{0},d)$ the equation with prescribed continuous boundary data has a solution in class $W^{2}_{p,\text{loc}}$ . Our results are new even if $H$ is linear.

On diffusion processes with drift in $L_{d}$

We investigate properties of Markov quasi-diffusion processes corresponding to elliptic operators $L=a^{ij}D_{ij}+b^{i}D_{i}$ , acting on functions on $\mathbb{R}^{d}$ , with measurable coefficients, bounded and uniformly elliptic $a$ and $b\in L_{d}(\mathbb{R}^{d})$ . We show that each of them is strong Markov with strong Feller transition semigroup $T_{t}$ , which is also a continuous bounded semigroup in $L_{d_{0}}(\mathbb{R}^{d})$ for some $d_{0}\in (d/2, d)$ . We show that $T_{t}$ , $t>0$ , has a kernel $p_{t}(x,y)$ which is summable in $y$ to the power of $d_{0}/(d_{0}-1)$ . This leads to the parabolic Aleksandrov estimate with power of summability $d_{0}$ instead of the usual $d+1$ . For the probabilistic solutions, associated with such a process, of the problem $Lu=f$ in a bounded domain $D\subset\mathbb{R}^{d}$ with boundary condition $u=g$ , where $f\in L_{d_{0}}(D)$ and $g$ is bounded, we show that it is Hölder continuous. Parabolic version of this problem is treated as well. We also prove Harnack's inequality for harmonic and caloric functions associated with such a process. Finally, we show that the probabilistic solutions are $L_{d_{0}}$ -viscosity solutions.

On stochastic equations with drift in $L_{d}$

For Itô stochastic equations in $\mathbb{R}^{d}$ with drift in $L_{d}$ several results are discussed such as the existence of weak solutions, the existence of the corresponding Markov process, Aleksandrov type estimates of their Green's functions, which yield their summability to the power of $d/(d-1)$ , the Fabes-Stroock type estimates which show that Green's functions are summable to a higher degree, the Fanghua Lin type estimates, which are one of the main tools in the $W^{2}_{p}$ -theory of fully nonlinear elliptic equations, the fact that Green's functions are in the class $A_{\infty}$ of Muckenhoupt and a few other results.

On stochastic Itô processes with drift in $L_{d}$

For Itô stochastic processes in $\mathbb{R}^{d}$ with drift in $L_{d}$ Aleksandrov's type estimates are established in the elliptic and parabolic settings. They are applied to estimating the resolvent operators of the corresponding elliptic and parabolic operators in $L_{p}$ and $L_{p+1}$ , respectively, where $p\geq d$ .

On strong solutions of Itô's equations with a $\,\in W^{1}_{d}$ and b $\,\in L_{d}$

We consider Itô uniformly nondegenerate equations with time independent coefficients, the diffusion coefficient in $W^{1}_{d,loc}$ , and the drift in $L_{d}$ . We prove the unique strong solvability for any starting point and prove that as a function of the starting point the solutions are Hölder continuous with any exponent $<1$ . We also prove that if we are given a sequence of coefficients converging in an appropriate sense to the original ones, then the solutions of approximating equations converge to the solution of the original one.

Rubio de Francia extrapolation theorem and related topics in the theory of elliptic and parabolic equations. A survey

We give a brief overview of the history of the Sobolev mixed norm theory of linear elliptic and parabolic equations and the recent development in this theory based on the Rubio de Francia extrapolation theorem. A self contained proof of this theorem along with other relevant tools of Real Analysis are also presented as well as an application to mixed norm estimates for fully nonlinear equations.

preprint2016arXiv

On an energy equality in the theory of evolution equations

We give an elementary proof of a basic "energy" equality in the theory of evolution equations in the setting, which usually starts with introducing $V\subset H=H^{*}\subset V^{*}$ .

preprint2016arXiv

On nonequivalence of regular boundary points for second-order elliptic operators

In this paper we present examples of nondivergence form second order elliptic operators with continuous coefficients such that $L$ has an irregular boundary point that is regular for the Laplacian. Also for any eigenvalue spread <1 of the matrix of the coefficients we provide an example of operator with discontinuous coefficients that has regular boundary points nonequivalent to Laplacian's (we give examples for each direction of nonequivalence). All examples are constructed for each dimension starting with 3.

preprint2016arXiv

On the existence of $W^{2}_{p}$ solutions for fully nonlinear elliptic equations under either relaxed or no convexity assumptions

We establish the existence of solutions of fully nonlinear elliptic second-order equations like $H(v,Dv,D^{2}v,x)=0$ in smooth domains without requiring $H$ to be convex or concave with respect to the second-order derivatives. Apart from ellipticity nothing is required of $H$ at points at which $|D^{2}v|\leq K$ , where $K$ is any given constant. For large $|D^{2}v|$ some kind of relaxed convexity assumption with respect to $D^{2}v$ mixed with a VMO condition with respect to $x$ are still imposed. The solutions are sought in Sobolev classes. We also establish the solvability without almost any conditions on $H$ , apart from ellipticity, but of a "cut-off" version of the equation $H(v,Dv,D^{2}v,x)=0$ .

preprint2015arXiv

On parabolic equations in one space dimension

Several negative results are presented concerning the solvability in Sobolev classes of the Cauchy problem for the inhomogeneous second-order uniformly parabolic equations without lower order terms in one space dimension. The main coefficient is assumed to be a bounded measurable function of $(t,x)$ bounded away from zero. We also discuss upper and lower estimates of certain kind on the fundamental solutions of such equations.

Approximating the value functions for stochastic differential games with the ones having bounded second derivatives

We show a method of uniform approximation of the value functions of uniformly nondegenerate stochastic differential games in smooth domains up to a constant over $K$ with the ones having second-order derivatives bounded by a constant times $K$ for any $K\geq1$

Hörmander's theorem for stochastic partial differential equations

We prove Hörmander's type hypoellipticity theorem for stochastic partial differential equations when the coefficients are only measurable with respect to the time variable. The need for such kind of results comes from filtering theory of partially observable diffusion processes, when even if the initial system is autonomous, the observation process enters the coefficients of the filtering equation and makes them time-dependent with no good control on the smoothness of the coefficients with respect to the time variable.

On regularity properties and approximations of value functions for stochastic differential games in domains

We prove that for any constant $K\geq1$ , the value functions for time homogeneous stochastic differential games in the whole space can be approximated up to a constant over $K$ by value functions whose second-order derivatives are bounded by a constant times $K$ . On the way of proving this result we prove that the value functions for stochastic differential games in domains and in the whole space admit estimates of their Lipschitz constants in a variety of settings.

On the independence of the value function for stochastic differential games of the probability space

We show that the value function in a stochastic differential game does not change if we keep the same space $(Ω,\mathcal{F})$ but introduce probability measures by means of Girsanov's transformation {\em depending} on the policies of the players. We also show that the value function does not change if we allow the driving Wiener processes to depend on the policies of the players. Finally, we show that the value function does not change if we perform a random time change with the rate depending on the policies of the players.

On the rate of convergence of finite-difference approximations for elliptic Isaacs equations in smooth domains

We show that the there exists an algebraic rate of convergence of solutions of finite-difference approximations for uniformly elliptic Isaacs in smooth bounded domains.

To the theory of viscosity solutions for uniformly elliptic Isaacs equations

We show how a theorem about solvability in $C^{1,1}$ of special Isaacs equations can be used to obtain existence and uniqueness of viscosity solutions of general uniformly nondegenerate Isaacs equations. We apply it also to establish the $C^{1+χ}$ regularity of viscosity solutions and show that finite-difference approximations have an algebraic rate of convergence. The main coefficients of the Isaacs equations are supposed to be in $C^γ$ with $γ$ slightly less than $1/2$ .

To the theory of viscosity solutions for uniformly parabolic Isaacs equations

We show how a theorem about the solvability in $W^{1,2}_{\infty}$ of special parabolic Isaacs equations can be used to obtain the existence and uniqueness of viscosity solutions of general uniformly nondegenerate parabolic Isaacs equations. We apply it also to establish the $C^{1+χ}$ regularity of viscosity solutions and show that finite-difference approximations have an algebraic rate of convergence. The main coefficients of the Isaacs equations are supposed to be in $C^γ$ with respect to the spatial variables with $γ$ slightly less than $1/2$ .

An ersatz existence theorem for fully nonlinear parabolic equations without convexity assumptions

We show that for any uniformly parabolic fully nonlinear second-order equation with bounded measurable "coefficients" and bounded "free" term in the whole space or in any cylindrical smooth domain with smooth boundary data one can find an approximating equation which has a continuous solution with the first and the second spatial derivatives under control: bounded in the case of the whole space and locally bounded in case of equations in cylinders. The approximating equation is constructed in such a way that it modifies the original one only for large values of the second spatial derivatives of the unknown function. This is different from a previous work of Hongjie Dong and the author where the modification was done for large values of the unknown function and its spatial derivatives.

Hörmander's theorem for parabolic equations with coefficients measurable in the time variable

We are dealing with possibly degenerate second-order parabolic operators whose coefficients are infinitely differentiable with respect to space variables and only measurable with respect to the time variable. We impose the Hörmander condition on the diffusion coefficients and prove that the solutions of the corresponding equations with right-hand sides which are infinitely differentiable in the space variables in a space-time domain have also this property.

Hypoellipticity for filtering problems of partially observable diffusion processes

We prove that under Hörmander's type conditions on the coefficients of the unobservable component of a partially observable diffusion process the filtering density is infinitely differentiable and can be represented as the integral of an infinitely differentiable kernel against the prior initial distribution. These results are derived from more general results obtained for SPDEs. One the main novelty of the paper is the existence and smoothness of the kernel, another that we allow the coefficients of our partially observable process to be just measurable with respect to the time variable.

On a result of Jessica Lin

In Theorem 1.5 of [5] an estimate from below is given for nonnegative supersolutions of parabolic equations in terms of the measure of the set where the right-hand side is less than $-1$ . The given proof contains many interesting details which are certainly useful in many other situations. However, as long as the statement of Theorem 1.5 of \cite{JL} is concerned, it can be proved in a much shorter way, which we present here.

On singularity as a function of time of a conditional distribution of an exit time

We establish the singularity with respect to Lebesgue measure as a function of time of the conditional probability that the sum of two one-dimensional Brownian motions will exit from the unit interval before time $t$ , given the trajectory of the second Brownian motion up to the same time. On the way of doing so we show that if one solves the one-dimensional heat equation with zero condition on a trajectory of a one-dimensional Brownian motion, which is the lateral boundary, then for each moment of time with probability one the normal derivative of the solution is zero, provided that the diffusion of the Brownian motion is sufficiently large.

A relatively short proof of Itô's formula for SPDEs and its applications

We give a short proof of Itô's formula for stochastic Hilbert-space valued processes in the setting $V\subset H\subset V^{*}$ based on the possibility to lift the stochastic differentials, which are originally in $V^{*}$ , into $H$ . Using this result we also prove the maximum principle for second-order SPDEs in arbitrary domains.

On $C^{1+α}$ regularity of solutions of Isaacs parabolic equations with VMO coefficients

We prove that boundary value problems for fully nonlinear second-order parabolic equations admit $L_{p}$ -viscosity solutions, which are in $C^{1+α}$ for an $α\in(0,1)$ . The equations have a special structure that the "main" part containing only second-order derivatives is given by a positive homogeneous function of second-order derivatives and as a function of independent variables it is measurable in the time variable and, so to speak, VMO in spatial variables.

On the dynamic programming principle for uniformly nondegenerate stochastic differential games in domains

We prove the dynamic programming principe for uniformly nondegenerate stochastic differential games in the framework of time-homogeneous diffusion processes considered up to the first exit time from a domain. The zeroth-order "coefficient" and the "free" term are only assumed to be measurable. In contrast with previous results established for constant stopping times we allow arbitrary stopping times and randomized ones as well. The main assumption, which will be removed in a subsequent article, is that there exists a sufficiently regular solution of the Isaacs equation.

On the dynamic programming principle for uniformly nondegenerate stochastic differential games in domains and the Isaacs equations

We prove the dynamic programming principe for uniformly nondegenerate stochastic differential games in the framework of time-homogeneous diffusion processes considered up to the first exit time from a domain. In contrast with previous results established for constant stopping times we allow arbitrary stopping times and randomized ones as well. There is no assumption about solvability of the the Isaacs equation in any sense (classical or viscosity). The zeroth-order "coefficient" and the "free" term are only assumed to be measurable in the space variable. We also prove that value functions are uniquely determined by the functions defining the corresponding Isaacs equations and thus stochastic games with the same Isaacs equation have the same value functions.

On the existence of $W^{2}_{p}$ solutions for fully nonlinear elliptic equations under relaxed convexity assumptions

We establish the existence and uniqueness of solutions of fully nonlinear elliptic second-order equations like $H(v,Dv,D^{2}v,x)=0$ in smooth domains without requiring $H$ to be convex or concave with respect to the second-order derivatives. Apart from ellipticity nothing is required of $H$ at points at which $|D^{2}v|\leq K$ , where $K$ is any given constant. For large $|D^{2}v|$ some kind of relaxed convexity assumption with respect to $D^{2}v$ mixed with a VMO condition with respect to $x$ are still imposed. The solutions are sought in Sobolev classes.

On the existence of smooth solutions for fully nonlinear elliptic equations with measurable "coefficients" without convexity assumptions

We show that for any uniformly elliptic fully nonlinear second-order equation with bounded measurable "coefficients" and bounded "free" term one can find an approximating equation which has a unique continuous and having the second derivatives locally bounded solution in a given smooth domain with smooth boundary data. The approximating equation is constructed in such a way that it modifies the original one only for large values of the unknown function and its derivatives.

Rate of convergence of difference approximations for uniformly nondegenerate elliptic Bellman's equations

We show that the rate of convergence of solutions of finite-difference approximations for uniformly elliptic Bellman's equations is of order at least $h^{2/3}$ , where $h$ is the mesh size. The equations are considered in smooth bounded domains.