arXiv subjects

Jingtao Shi

Publications and source records attributed to Jingtao Shi.

At least 37 records · Page 2Linked to original sources

arXiv2025

Global Maximum Principle for Partially Observed Risk-Sensitive Progressive Optimal Control of FBSDE with Poisson Jumps

This paper is concerned with one kind of partially observed progressive optimal control problems of coupled forward-backward stochastic systems driven by both Brownian motion and Poisson random measure with risk-sensitive criteria. The control domain is not necessarily convex, and the control variable can enters into all the coefficients. The observation equation also has correlated noises with the state equation. Under the Poisson jump setting, the original problem is equivalent to a complete information stochastic recursive optimal control problem of a forward-backward system with quadratic-exponential generator. In order to establish the first- and second-order variations, some new techniques are introduced to overcome difficulties caused by the quadratic-exponential feature. A new global stochastic maximum principle is deduced. As an application, a risk-sensitive optimal investment problem with factor model is studied. Moreover, the risk-sensitive stochastic filtering problem is also studied, which involves both Brownian and Poissonian correlated noises. A modified Zakai equation is obtained.

math.OC↗

arXiv2025

A General Maximum Principle for Progressive Optimal Control of Fully Coupled Forward-Backward Stochastic Systems with Jumps

This paper is concerned with a general maximum principle for the fully coupled forward-backward stochastic optimal control problem with jumps, where the control domain is not necessarily convex, within the progressively measurable framework. A distinct feature in this paper is that the solution $Z$ of BSDEPs could include the variable ``$e$'', further, the diffusion term of BSDEPs takes the form $\int_{\mathcal{E}}Z_{(t,e)}ν(d e)d W_t$ rather than the conventional $Z_t dW_t$, reflecting the essential coupling between the solution component $Z$ and the Polish space $\mathcal{E}$.

math.OC↗

arXiv2025

Linear-Quadratic Mean Field Stackelberg Stochastic Differential Game with Partial Information and Common Noise

This paper is concerned with a linear-quadratic mean field Stackelberg stochastic differential game with partial information and common noise, which contains a leader and a large number of followers. To be specific, the followers face a large population Nash game after the leader first announces his strategy, while the leader will then optimize his own cost functional on consideration of the followers' reactions. The state equation of the leader and followers are both general stochastic differential equations, where the diffusion terms contain both the control and state variables. However, the followers' average state terms enter into the drift term of the leader's state equation, reflecting that the leader's state is influenced by the followers' states. By virtue of stochastic maximum principle with partial information and optimal filter technique, we deduce the open-loop adapted decentralized strategies and feedback decentralized strategies of this leader-followers system, and demonstrate that the decentralized strategies are the corresponding $\varepsilon$-Stackelberg-Nash equilibrium.

math.OC↗

arXiv2025

The Optimal Control Problem of Fully Coupled FBSDEs Driven by Sub-diffusion with Applications

This paper is devoted to an optimal control problem of fully coupled forward-backward stochastic differential equations driven by sub-diffusion, whose solutions are not Markov processes. The stochastic maximum principle is obtained, where the control domain may not be convex and the diffusion term is independent of the control variable. Additionally, problem with state constraint is researched by using Ekeland's variational principle. The theoretical results obtained are applied to a cash management optimization problem in bear market, and the optimal strategy is derived.

math.OC↗

arXiv2025

A Stochastic Linear-Quadratic Leader-Follower Differential Game with Elephant Memory

This paper is concerned with a stochastic linear-quadratic leader-follower differential game with elephant memory. The model is general in that the state equation for both the leader and the follower includes the elephant memory of the state and the control, which are part of the diffusion term. Under certain assumptions, the state feedback representation of the open-loop Stackelberg strategy is derived by introducing two Riccati equations and a special matrix-valued equation. Finally, theoretical results are illustrated by means of an example concerning a dynamic advertising problem with elephant memory.

math.OC↗

arXiv2025

Decentralized Strategies for Backward Linear-Quadratic Mean Field Games and Teams

This paper studies a new class of linear-quadratic mean field games and teams problem, where the large-population system satisfies a class of $N$ weakly coupled linear backward stochastic differential equations (BSDEs), and $z_i$ (a part of solution of BSDE) enter the state equations and cost functionals. By virtue of stochastic maximum principle and optimal filter technique, we obtain a Hamiltonian system first, which is a fully coupled forward-backward stochastic differential equation (FBSDE). Decoupling the Hamiltonian system, we derive a feedback form optimal strategy by introducing Riccati equations, stochastic differential equation (SDE) and BSDE. Finally, we provide a numerical example to simulate our results.

math.OC↗

arXiv2024

Linear-Quadratic Stackelberg Mean Field Games and Teams with Arbitrary Population Sizes

This paper addresses a linear-quadratic Stackelberg mean field (MF) games and teams problem with arbitrary population sizes, where the game among the followers is further categorized into two types: non-cooperative and cooperative, and the number of followers can be finite or infinite. The leader commences by providing its strategy, and subsequently, each follower optimizes its individual cost or social cost. A new de-aggregation method is applied to solve the problem, which is instrumental in determining the optimal strategy of followers to the leader's strategy. Unlike previous studies that focus on MF games and social optima, and yield decentralized asymptotically optimal strategies relative to the centralized strategy set, the strategies presented here are exact decentralized optimal strategies relative to the decentralized strategy set. This distinction is crucial as it highlights a shift in the approach to MF systems, emphasizing the precision and direct applicability of the strategies to the decentralized context. In the wake of the implementation of followers' strategies, the leader is confronted with an optimal control problem driven by high-dimensional forward-backward stochastic differential equations (FBSDEs). By variational analysis, we obtain the decentralized strategy for the leader. By applying the de-aggregation method and employing dimension expansion to decouple the high-dimensional FBSDEs, we are able to derive a set of decentralized Stackelberg-Nash or Stackelberg-team equilibrium solution for all players.

math.OC↗

arXiv2024

Three-Level Multi-Leader-Follower Incentive Stackelberg Differential Game with $H_\infty$ Constraint

This paper is concerned with a three-level multi-leader-follower incentive Stackelberg game with $H_\infty$ constraint. Based on $H_2/H_\infty$ control theory, we firstly obtain the worst-case disturbance and the team-optimal strategy by dealing with a nonzero-sum stochastic differential game. The main objective is to establish an incentive Stackelberg strategy set of the three-level hierarchy in which the whole system achieves the top leader's team-optimal solution and attenuates the external disturbance under $H_\infty$ constraint. On the other hand, followers on the bottom two levels in turn attain their state feedback Nash equilibrium, ensuring incentive Stackelberg strategies while considering the worst-case disturbance. By convex analysis theory, maximum principle and decoupling technique, the three-level incentive Stackelberg strategy set is obtained. Finally, a numerical example is given to illustrate the existence of the proposed strategy set.

math.OC↗

arXiv2024

Backward Linear-Quadratic Mean Field Stochastic Differential Games: A Direct Method

This paper studies a linear-quadratic mean-field game of stochastic large-population system, where the large-population system satisfies a class of $N$ weakly coupled linear backward stochastic differential equation. Different from the fixed-point approach commonly used to address large population problems, we first directly apply the maximum principle and decoupling techniques to solve a multi-agent problem, obtaining a centralized optimal strategy. Then, by letting $N$ tend to infinity, we establish a decentralized optimal strategy. Subsequently, we prove that the decentralized optimal strategy constitutes an $ε$-Nash equilibrium for this game. Finally, we provide a numerical example to simulate our results.

math.OC↗

arXiv2024

Stochastic Linear-Quadratic Stackelberg Differential Game with Asymmetric Informational Uncertainties: Robust Optimization Approach

This paper is concerned with a two-person zero-sum indefinite stochastic linear-quadratic Stackelberg differential game with asymmetric informational uncertainties, where both the leader and follower face different and unknown disturbances. We take a robust optimization approach and soft-constraint analysis, a min-max stochastic linear-quadratic optimal control problem is solved by the follower firstly. Then, the leader deal with a max-min stochastic linear-quadratic optimal control problem of forward-backward stochastic differential equations in an augmented space. State feedback representation of the robust Stackelberg equilibrium is given in a more explicit form by decoupling technique, via some Riccati equations.

math.OC↗

arXiv2024

Direct Approach of Indefinite Linear-Quadratic Mean Field Games

This paper is concerned with an indefinite linear-quadratic mean field games of stochastic large-population system, where the individual diffusion coefficients can depend on both the state and the control of the agents. Moreover, the control weights in the cost functionals could be indefinite. A direct approach is used to derive the $ε$-Nash equilibrium strategy. First, we formally solving an $N$-player game problem within a vast and finite population setting. Subsequently, decoupling or reducing high-dimensional systems by introducing two Riccati equations explicitly yields centralized strategies, contingent on the state of a specific player and the average state of the population. As the population size $N$ goes infinity, the construction of decentralized strategies becomes feasible. Then, we demonstrated they are an $ε$-Nash equilibrium. Numerical examples are provided to demonstrate the effectiveness of the proposed strategies.

math.OC↗

arXiv2024

Direct Approach of Linear-Quadratic Stackelberg Mean Field Games of Backward-Forward Stochastic Systems

This paper is concerned with a linear-quadratic (LQ) Stackelberg mean field games of backward-forward stochastic systems, involving a backward leader and a substantial number of forward followers. The leader initiates by providing its strategy, and subsequently, each follower optimizes its individual cost. A direct approach is applied to solve this game. Initially, we address a mean field game problem, determining the optimal response of followers to the leader's strategy. Following the implementation of followers' strategies, the leader faces an optimal control problem driven by high-dimensional forward-backward stochastic differential equations (FBSDEs). Through the decoupling of the high-dimensional Hamiltonian system using mean field approximations, we formulate a set of decentralized strategies for all players, demonstrated to be an $(ε_1, ε_2)$-Stackelberg equilibrium.

math.OC↗

arXiv2024

Relationship between General MP and DPP for the Stochastic Recursive Optimal Control Problem With Jumps: Viscosity Solution Framework

This paper is concerned with the relationship between general maximum principle and dynamic programming principle for the stochastic recursive optimal control problem with jumps, where the control domain is not necessarily convex. Relations among the adjoint processes, the generalized Hamiltonian function and the value function are proved, under the assumption of a smooth value function and within the framework of viscosity solutions, respectively. Some examples are given to illustrate the theoretical results.

math.OC↗

arXiv2024

An overlapping information linear-quadratic Stackelberg stochastic differential game with two leaders and two followers

This paper is concerned with an overlapping information linear-quadratic (LQ) Stackelberg stochastic differential game with two leaders and two followers, where the diffusion terms of the state equation contain both the control and state variables. A distinct feature lies in that, the noisy information available to the leaders and the followers may be asymmetric and have overlapping part. Using a coupled system of Riccati equations, the followers first solve an LQ nonzero-sum stochastic differential Nash game with partial information, and then the leaders solve a partial information LQ nonzero-sum stochastic differential Nash game driven by a conditional mean-field type forward-backward stochastic differential equation (CMF-FBSDE). By maximum principle, completion of squares and decoupling methods, the state-estimate feedback representation of the Stackelberg-Nash equilibrium is obtained.

math.OC↗

arXiv2023

Closed-Loop Solvability of Linear Quadratic Mean-Field Type Stackelberg Stochastic Differential Games

This paper is devoted to a Stackelberg stochastic differential game for a linear mean-field type stochastic differential system with a mean-field type quadratic cost functional in finite horizon. The coefficients in the state equation and weighting matrices in the cost functional are all deterministic. Closed-loop Stackelberg equilibrium strategies are introduced which require to be independent of initial states. Follower's problem is solved firstly, which is a stochastic linear quadratic optimal control problem. By converting the original problem into a new one whose optimal control is known, the closed-loop optimal strategy of the follower is characterized by two coupled Riccati equations as well as a linear mean-field type backward stochastic differential equation. Then the leader turns to solve a stochastic linear quadratic optimal control problem for a mean-field type forward-backward stochastic differential equation. Necessary conditions for the existence of closed-loop optimal strategies for the leader is given by the existence of two coupled Riccati equations with a linear mean-field type backward stochastic differential equation. The solvability of Riccati equations of leader's optimization problem is discussed in the case where the diffusion term of the state equation does not contain the control process of the follower. Moreover, leader's value function is expressed via two backward stochastic differential equations and two Lyapunov equations.

math.OC↗

arXiv2023

A Risk-Sensitive Global Maximum Principle for Controlled Fully Coupled FBSDEs with Applications

This paper is concerned with a kind of risk-sensitive optimal control problem for fully coupled forward-backward stochastic systems. The control variable enters the diffusion term of the state equation and the control domain is not necessarily convex. A new global maximum principle is obtained without assuming that the value function is smooth. The maximum condition, the first- and second-order adjoint equations heavily depend on the risk-sensitive parameter. An optimal control problem with a fully coupled linear forward-backward stochastic system and an exponential-quadratic cost functional is discussed. The optimal feedback control and optimal cost are obtained by using Girsanov's theorem and completion-of-squares approach via risk-sensitive Riccati equations. A local solvability result of coupled risk-sensitive Riccati equations is given by Picard-Lindel$\ddot{o}$f's Theorem.

math.OC↗

arXiv2023

A general maximum principle for optimal control of stochastic differential delay systems

In this paper, we solve an open problem and obtain a general maximum principle for a stochastic optimal control problem where the control domain is an arbitrary non-empty set and all the coefficients (especially the diffusion term and the terminal cost) contain the control and state delay. In order to overcome the difficulty of dealing with the cross term of state and its delay in the variational inequality, we propose a new method: transform a delayed variational equation into a Volterra integral equation without delay, and introduce novel first-order, second-order adjoint equations via the backward stochastic Volterra integral equation theory. Finally we express these two kinds of adjoint equations in more compact anticipated backward stochastic differential equation types for several special yet typical control systems.

math.OC↗

arXiv2022

Mixed Leadership Stochastic Differential Game in Feedback Information Pattern with Applications

This paper is devoted to a high-dimensional mixed leadership stochastic differential game on a finite horizon in feedback information mode, where the control variables enter into the diffusion term of state equation. A verification theorem for the feedback Stackelberg-Nash equilibrium is obtained by using a system of coupled and fully nonlinear parabolic partial differential equations. We apply the verification theorem to deal with a dynamic innovation and pricing decision problem where the buyer acts as the leader in the pricing decisions and the dynamic model is stochastic. Via the solutions of coupled Riccati equations, we explicitly express the feedback equilibrium strategies of innovation and pricing. And by analysis, the local existence and uniqueness of the solutions of the coupled Riccati equations is derived. We also conduct some numerical analyses to discuss the effects of model parameters on the feedback equilibrium strategies.

math.OC↗