Abstract:We derive an Itô-type formula for a measure-valued process that has a decomposition analogous to a classical semimartingale. The derivation begins with a time partitioning approach similar to the classical proof of Itô's formula. To address the new challenges arising from the measure-valued setting, we employ symmetric polynomials to approximate the second-order linear derivative of the functional on finite measures, alongside certain localization techniques. A controlled superprocess with a binary branching mechanism can be interpreted as a weak solution to a controlled stochastic partial differential equation (SPDE), which naturally leads to such a decomposition. Consequently, this Itô-type formula makes it possible to derive the Hamilton-Jacobi-Bellman (HJB) equation and the verification theorem for controlled superprocesses with a binary branching mechanism. Additionally, we propose a heuristic definition for the viscosity solution of an equation involving derivatives on finite measures. We prove that a continuous value function is a viscosity solution in this sense and demonstrate the uniqueness of the viscosity solution when the second-order derivative term on the measure vanishes.

What problem does this paper attempt to address?

The core problem that this paper attempts to solve is how to establish an Itô - type formula for some measure - valued processes with a decomposition similar to the classical semimartingale. Specifically, the author aims to solve this problem through the following points: 1. **Establishing the Itô - type formula**: Establish an Itô - type formula for specific measure - valued processes, which can be represented as solutions to stochastic partial differential equations (SPDEs) driven by martingale measures. This formula will help to derive the generalized martingale problem that governs super - processes and provide a new method for deriving the Hamilton - Jacobi - Bellman (HJB) equation and the verification theorem. 2. **Handling new challenges**: When proving this Itô - type formula, the author encountered some new challenges. For example, when using Taylor expansion to approximate the linear derivative, the sum of the second - order terms is no longer a discrete quadratic variation. To solve this problem, the author used a linear combination of products of univariate functions to approximate the second - order linear derivative and combined with the localization technique to strictly prove the rationality of the approximation process. 3. **Applying to control super - processes**: Through the above Itô - type formula, the author was able to derive the HJB equation and the verification theorem for controlling super - processes and prove that the value function is a viscosity solution of the HJB equation. In addition, the author also explored the potential applications of this type of Itô - type formula in the study of other measure - valued processes and SPDE systems. ### Specific problem description The main problem mentioned in the paper is to establish an Itô - type formula for the controlled super - process with a binary branching mechanism. This type of process can be regarded as a weak solution to an SPDE driven by space - time white noise. By introducing tools such as linear functional derivatives and Lions derivatives, the author overcame the limitations of traditional methods in dealing with finite measure spaces and proposed a new framework to handle these problems. ### Key points of the solution - **Itô - type formula for measure - valued processes**: Through the time - splitting method and symmetric polynomial approximation techniques, the author successfully established an Itô - type formula for measure - valued processes. - **Definition and uniqueness of viscosity solutions**: The author proposed a heuristic definition of viscosity solutions for derivative equations on finite measures and proved that the continuous value function is a viscosity solution in this sense, and when the second - order derivative term disappears, the viscosity solution is unique. - **Application of the dynamic programming principle**: Based on the dynamic programming principle, the author derived the HJB equation and the verification theorem for controlling super - processes, thus providing a theoretical basis for solving optimal control problems. In conclusion, this paper successfully solved the difficult problem of establishing an Itô - type formula for measure - valued processes by introducing new mathematical tools and techniques and applied it to the study of controlled super - processes.

An Itô-type formula for some measure-valued processes and its application on controlled superprocesses

Controlled superprocesses and HJB equation in the space of finite measures

Itô's Formula for Itô processes defined with respect to a cylindrical-martingale valued measure

On Itô formulas for jump processes

Controlled Measure-Valued Martingales: a Viscosity Solution Approach

A $C^1$-Itô's formula for flows of semimartingale distributions

Controlled Occupied Processes and Viscosity Solutions

Solving non-Markovian Stochastic Control Problems driven by Wiener Functionals

Itō and Itō-Wentzell chain rule for flows of conditional laws of continuous semimartingales: an easy approach

Itô's formula for noncommutative $C^2$ functions of free Itô processes

Characteristics and It{ô}'s formula for weak Dirichlet processes: an equivalence result

Occupied Processes: Going with the Flow

Itô's formula for finite variation Lévy processes: The case of non-smooth functions

Revisiting Stochastic Realization Theory using Functional Itô Calculus

Stable Central Limit Theorems for Super Ornstein-Uhlenbeck Processes

Minimal Kullback–Leibler Divergence for Constrained Lévy–Itô Processes

Control of Conditional Processes and Fleming--Viot Dynamics

Controlled Singular Volterra Integral Equations and Pontryagin Maximum Principle

Ito's formula for flows of conditional measures on semimartingales

Change of variable formulas for non-anticipative functionals on path space

A strong law of large numbers for super-stable processes