Searcharxiv⌕ Search

arXiv subjects

Shaolin Ji

Publications and source records attributed to Shaolin Ji.

At least 19 recordsLinked to original sources

Learning the Brownian Motion in Machine Learning for Parametric Parabolic PDEs: A Backward Quadrature Approach

We propose a backward quadrature learning (BQL) method for solving high dimensional parametric semilinear parabolic partial differential equations (PDEs). To the best of our knowledge, this work is the first to extend the applicability of BSDE-based stochastic numerical methods to this class of PDEs. Specifically, inspired by the approach of learning the random variables (LRV)~\cite{BeckerJentzenMullerWurstemberger2024}, BQL incorporates learnable quadrature for Brownian increments at each time level into the backward recursion of BSDEs, enabling more efficient layerwise approximation of local conditional expectations. Furthermore, BQL introduces a continuation surrogate to facilitate the transfer of value-function information from subsequent to preceding time levels, avoiding nested quadrature while controlling the propagation of local approximation errors. Our main theoretical contribution is the derivation of a global error bound that explicitly characterizes the decomposition of the total error, together with a proof of first-order mean-square convergence under suitable approximation and integrability assumptions. Finally, numerical experiments demonstrate that BQL achieves competitive solution accuracy while substantially reducing the amortized computational cost per query. For the HJB benchmark across the tested dimensions, with $Q=7$ parameter queries, BQL achieves computational cost reductions of approximately 40-fold and 4-fold compared with P$^2$INN~[2] and FBSNN~[3], respectively.

math.NA↗

Deep backward multistep high-order regression methods for high dimensional nonlinear parabolic partial differential equations

We propose a deep backward multistep high-order regression (MHDBR) method for semilinear parabolic partial differential equations through their backward stochastic differential equation representations. The method combines multistep time discretization with regression targets formed by conditional expectations. Under explicit regularity, starting-value, forward-approximation and network-moment assumptions, we prove an error bound separating time discretization, conditional Monte Carlo sampling and neural-network approximation. The analysis assumes exact population minimization; finite-training-sample generalization and optimization errors are not included. For generators independent of the gradient variable, suitably constrained GroupSort networks and sufficiently large conditional sample sizes give order $J$ in the stated nodal $L^2$ error norm. The required network sizes depend on dimension, so this result does not establish dimension-independent complexity. We present a reproducible time-discretization test satisfying the analytical assumptions and three broader neural-network benchmarks, whose relation to the assumptions is discussed explicitly.

stat.ML↗

Soft-Update MSA for Nonconvex Stochastic Optimal Control: Theory and Algorithms

For the method of successive approximations (MSA), a fundamental challenge is that the stochastic maximum principle (SMP) provides an optimality condition rather than a descent rule. To overcome this difficulty for nonconvex stochastic control problems, we develop a soft-update MSA (SU-MSA) with a second-order-adjoint correction of the Hamiltonian and a neural realization called DeepSMP. Both an explicit stepsize rule guaranteeing cost descent and the convergence of the algorithm are established. Moreover, the global SMP condition is asymptotically satisfied when the corrected Hamiltonian is convex, while in a special case the cost is further proved to converge to the global optimum. To extend SU-MSA to high dimensions, we develop DeepSMP using a differentiable smooth approximation of the corrected Hamiltonian and analyze the convergence of the resulting objective and its gradient. Numerical experiments demonstrate the effectiveness of the proposed framework.

math.OC↗

Flow Matching under Noisy Latent Structure: Beyond Exact Low-Dimensional Support

Flow Matching (FM) learns a velocity field whose ODE transports a simple source distribution to a target law. Existing finite-sample theory largely treats ambient-space regularity or data supported exactly on low-dimensional sets. We study linear FM under a noisy latent-generator model, where a low-dimensional Hölder map is perturbed by nondegenerate ambient Gaussian noise, so the target law is full-dimensional despite its latent structure. We construct a spatially regular ReLU velocity class and establish non-asymptotic high-probability approximation and estimation bounds whose leading sample-size exponent is governed by the latent dimension rather than the ambient dimension, with ambient and noise dependence kept explicit. Fixed positive target noise keeps the interpolation nondegenerate over the full time interval. The same spatial regularity propagates the learned velocity error through the transport ODE, yielding a corresponding Wasserstein convergence guarantee. These results show that exact low-dimensional support is not necessary for Flow Matching to retain latent-dimensional statistical behavior.

cs.LG↗

Variable-Coefficient Parabolic Equations and Finite-Horizon Hamilton--Jacobi--Bellman Equations in Augmented Spectral Barron Spaces

We establish a whole-space solution framework in augmented spectral Barron spaces for uniformly elliptic parabolic equations with state-dependent principal coefficients. A frozen-symbol parametrix yields bounded Green and terminal operators and a one-derivative smoothing estimate on arbitrary finite horizons, without requiring the spatial variation of the principal coefficient to be perturbatively small. We then apply this linear framework to a finite-horizon Hamilton--Jacobi--Bellman equation. A semi-explicit gradient iteration converges on sufficiently short horizons at a Gamma-factorial rate in a whole-space augmented Barron norm, and its limit is identified with the stochastic-control value function. Finally, temporal Jackson approximation and spatial spectral Barron approximation give joint shallow cosine-network approximations in space and time for the value function and the optimal feedback, with quantitative error and neuron-count bounds. Taken together, these results connect variable-coefficient parabolic well-posedness and smoothing in augmented spectral Barron spaces with nonlinear HJB solvability and quantitative neural-network approximation.

math.OC↗

Decoupled Probabilistic Forecasting and Arbitrage-Aware Refinement of Implied Volatility Surfaces

Implied volatility surface forecasting is essential for option valuation, hedging,and risk management, but remains difficult because future surfaces are stochastic while pricing inputs must satisfy static no-arbitrage shape restrictions. We propose a decoupled generative refinement framework for IVS forecasting as an operational risk surface modeling problem. The first stage uses a conditional diffusion model to learn the conditional distribution of future surfaces. The generated ensemble captures predictive distributional variation, and its median provides a robust representative surface for subsequent refinement. The second stage introduces a Surface Aware Attention Module (SAAM), a cross sectional refinement operator that improves fit to market observations and staticno-arbitrage diagnostics for the representative surface. This design separates distribution learning from surface refinement, allowing the diffusion model to capture stochastic market dynamics while SAAM controls static no-arbitrage residual violations on the final surface. We evaluate the framework on CSI 300 index options from June 2020 to September 2024 under daily and minute level forecasting protocols. The diffusion stage improves forecasting accuracy and produces predictive intervals that vary across moneyness, maturity, and sampling frequency. The refinement stage improves fitting accuracy against market observations and reduces measured static no-arbitrage residual violations, with stronger gains at the minute level. Attention diagnostics suggest that SAAM performs adaptive cross sectional refinement rather than fixed local smoothing

q-fin.CP↗

Finite horizon stochastic $H_2/H_\infty$ control for continuous-time mean-field systems with Poisson jumps

The stochastic $H_2/H_\infty$ control problem for continuous-time mean-field stochastic differential equations with Poisson jumps over finite horizon is investigated in this paper. Continuous and jump diffusion terms in the system depend not only on the state but also on the control input, external disturbance, and mean-field components. By employing the quasi-linear technique and the method of completing the square, a mean-field stochastic jump bounded real lemma of the system is derived, which plays a crucial role in solving stochastic $H_2/H_\infty$ control problem. It is demonstrated in this study that the feasibility of the stochastic $H_2/H_\infty$ control problem is equivalent to the solvability of four sets of cross-coupled generalized differential Riccati equations, thus generalizing the previous results to mean-field jump-diffusion systems. To validate the proposed methodology, a numerical simulation example is provided to illustrate the effectiveness of the control strategy. The results establish a systematic approach for designing $H_2/H_\infty$ controllers that simultaneously guarantee the robustness against disturbances and optimal performance for interacting particle systems.

math.OC↗

A deep backward regression-based scheme for high-dimensional nonlinear partial differential equations

We propose a deep backward regression-based (DBR) scheme for solving high-dimensional nonlinear parabolic partial differential equations. Building on the DBDP method of Huré, Pham, and Warin~\cite{HCPHWX20}, the proposed method reformulates the local backward losses through conditional expectations and trains the resulting regression problems sequentially in time. This conditional-expectation formulation replaces pathwise Brownian fluctuations in the Euler residual by their averaged effect and therefore provides an intrinsic variance-reduction mechanism before loss evaluation. In practice, the conditional expectations are approximated by local multi-path Monte Carlo averages, which leads to smoother training targets and improved numerical stability. Numerical experiments show that DBR performs competitively on standard high-dimensional benchmarks and is more stable than DBDP1 on the challenging unbounded benchmark considered in Example~2. Under an idealized population-loss minimization setting, we provide an error analysis and establish a half-order convergence result under suitable approximation and integrability assumptions. We also discuss an extension to variational inequalities.

math.NA↗

Global Convergence of Successive Approximations for Non-convex Stochastic Optimal Control Problems

This paper focuses on finding approximate solutions to stochastic optimal control problems with control domains being not necessarily convex, where the state trajectory is subject to controlled stochastic differential equations. The control-dependent diffusions make the traditional method of successive approximations (MSA) insufficient to reduce the value of cost functional in each iteration. Without adding extra terms over which to perform the Hamiltonian minimization, the MSA becomes sufficient by our novel error estimate involving a higher order backward adjoint equation. Under certain convexity assumptions on the coefficients (no convexity assumptions on the control domains), the value of the cost functional descends to the global minimum as the number of iterations tends to infinity. In particular, a convergence rate is available for a class of generalized linear-quadratic systems.

math.OC↗

A BSDE approach to the asymmetric risk-sensitive optimization and its applications

This paper is devoted to proposing a new asymmetric risk-sensitive criterion involving different risk attitudes toward varying risk sources. The criterion can only be defined through the initial value of the minimal solutions of quadratic backward stochastic differential equations (BSDEs). Before uncovering the mean-variance representation for the introduced criterion by the variational approach, some axioms are given for the first time to characterize a variance decomposition of square integrable random variables. The stochastic control problems under this criterion are described as a kind of stochastic recursive control problems that includes controlled quadratic BSDEs. An asymmetric risk-sensitive global stochastic maximum principle is derived when the quadratic BSDEs are equipped with bounded data. A closed-form solution of a stochastic linear-quadratic risk-sensitive control problem is obtained by introducing a novel completion-of-squares technique for controlled quadratic BSDEs. In addition, a dynamic portfolio optimization problem featuring a stochastic return rate is provided as an application of the asymmetric risk-sensitive control.

math.OC↗

Uniform Convergence Rate of the Nonparametric Estimator for Integrated Diffusion Processes

The nonparametric estimation of integrated diffusion processes has been extensively studied, with most existing research focusing on pointwise convergence. This paper is the first to establish uniform convergence rates for the Nadaraya-Watson estimators of their coefficients. We derive these rates over unbounded support under the assumptions of a vanishing observation interval and a long time horizon. Our findings serve as essential tools for specification testing and semiparametric inference in various diffusion models and time series, facilitating applications in finance, geology, and physics through nonparametric estimation methods.

math.ST↗

A deep learning method for solving stochastic optimal control problems driven by fully-coupled FBSDEs

In this paper,we mainly focus on the numerical solution of high-dimensional stochastic optimal control problem driven by fully-coupled forward-backward stochastic differential equations (FBSDEs in short) through deep learning. We first transform the problem into a stochastic Stackelberg differential game problem (leader-follower problem), then a bi-level optimization method is developed where the leader's cost functional and the follower's cost functional are optimized alternatively via deep neural networks. As for the numerical results, we compute two examples of the investment-consumption problem solved through stochastic recursive utility models, and the results of both examples demonstrate the effectiveness of our proposed algorithm.

math.OC↗

Nonparametric estimation of FBSDEs with random terminal time

This paper investigates the nonparametric estimation of the functional coefficients of the FBSDEs with random terminal time, including the local constant and local linear estimators. We provide complete two-dimensional asymptotics in both the time span and the sampling interval, allowing for the precise characterization of their distribution. Moreover, the empirical likelihood (EL) method to construct the data-driven confidence intervals for these estimators is provided. Some numerical simulations investigate the finite-sample properties of the estimators and compare the performance of the EL method and the conventional method in constructing confidence intervals based on asymptotic normality.

math.ST↗

A Stochastic Maximum Principle for Forward-backward Stochastic Control Systems with Quadratic Generators and Sample-wise Constraints

This paper examines the stochastic maximum principle (SMP) for a forward-backward stochastic control system where the backward state equation is characterized by the backward stochastic differential equation (BSDE) with quadratic growth and the forward state at the terminal time is constrained in a convex set with probability one. With the help of the theory of BSDEs with quadratic growth and the bounded mean oscillation (BMO) martingales, we employ the terminal perturbation approach and Ekeland's variational principle to obtain a dynamic stochastic maximum principle. The main result has a wide range of applications in mathematical finance and we investigate a robust recursive utility maximization problem with bankruptcy prohibition as an example.

math.OC↗

Mean-variance portfolio selection with nonlinear wealth dynamics and random coefficients

This paper studies the continuous time mean-variance portfolio selection problem with one kind of non-linear wealth dynamics. To deal the expectation constraint, an auxiliary stochastic control problem is firstly solved by two new generalized stochastic Riccati equations from which a candidate portfolio in feedback form is constructed, and the corresponding wealth process will never cross the vertex of the parabola. In order to verify the optimality of the candidate portfolio, the convex duality (requires the monotonicity of the cost function) is established to give another more direct expression of the terminal wealth level. The variance-optimal martingale measure and the link between the non-linear financial market and the classical linear market are also provided. Finally, we obtain the efficient frontier in closed form. From our results, people are more likely to invest their money in riskless asset compared with the classical linear market.

q-fin.MF↗

Maximum principle for discrete-time stochastic optimal control problem under distribution uncertainty

In this paper, we study a discrete-time stochastic optimal control problem under distribution uncertainty with convex control domain. By weak convergence method and Sion's minimax theorem, we obtain the variational inequality for cost functional under a reference probability $P^{\ast}$. Moreover, under the square integrability condition for noise and control, we establish the discrete-time stochastic maximum principle under $P^{\ast}$. Finally, we introduce a backward algorithm to calculate the reference probability $P^{\ast}$ and the optimal control $u^{\ast}$.

math.OC↗

A Modified Method of Successive Approximations for Stochastic Recursive Optimal Control Problems

Based on the stochastic maximum principle for the partially coupled forward-backward stochastic control system (FBSCS for short), a modified method of successive approximations (MSA for short) is established for stochastic recursive optimal control problems. The second-order adjoint processes are introduced in the augmented Hamiltonian minimization step since the control domain is not necessarily convex. Thanks to the theory of bounded mean oscillation martingales (BMO martingales for short), we give a delicate proof of the error estimate and then prove the convergence of the modified MSA algorithm. In a special case, we obtain a logarithmic convergence rate. When the control domain is convex and compact, a sufficient condition which makes the control returned from the MSA algorithm be a near-optimal control is given for a class of linear FBSCSs.

math.OC↗