Searcharxiv⌕ Search

arXiv subjects

Chuchu Chen

Publications and source records attributed to Chuchu Chen.

At least 19 recordsLinked to original sources

Explicit geometric integrators for Langevin equations with superquadratic potentials

We propose a new class of explicit methods for Langevin equations with a superquadratic potential that simultaneously preserve the conformal symplecticity and ergodicity as well as achieving high-order convergence. The construction is based on a mollified potential approach, in which the original potential is truncated to quadratic growth at infinity and globally mollified to control the superquadratic behavior while maintaining the underlying properties of the original dynamics. By appropriately choosing the cut-off radius, the far-field quadratic coefficient, and the step size, we obtain the coercivity of the mollified potential and derive its derivative estimates. These estimates allow us to prove uniform-in-time weak convergence for the proposed (including higher-order) explicit conformal symplectic numerical methods. Since the mollified potential is identical to the original one in a ball whose radius grows as the step size decreases, the computational cost of the new methods is essentially the same as that of the usual schemes (e.g., BAOAB) applied to the original potential. Numerical experiments are provided to illustrate the theoretical results.

math.NA↗

Almost sure asymptotics of numerical errors for SDEs with additive noise

In this paper, we characterize the sharp almost sure asymptotics for the numerical error of the Euler--Maruyama method applied to stochastic differential equations with additive noise. Specifically, we establish, to our knowledge, the first sharp laws of the iterated logarithm for Euler--Maruyama errors in this setting along geometric mesh refinements $(N_i)_{i\in\mathbb N^+}=(α^i)_{i\in\mathbb N^+}$ for any integer $α\geq 2$. These results identify $(N_i)^{-1}\sqrt{\log\log N_i}$ as the sharp almost sure error scale and provide an explicit expression for the path-dependent constants governing the asymptotic error envelopes. The LIL constants also admit a natural interpretation through the asymptotic error distribution, specifically through the conditional covariance matrix of the stable limit of the rescaled error given the driving Wiener process. A central challenge in establishing these results is to characterize the almost sure fluctuations of Skorohod integrals arising in the error expansion, which are generally anticipating, non-Gaussian, and dependent across refinement levels. The analysis also provides a framework for studying analogous asymptotics for other numerical methods.

math.NA↗

Weak convergence orders for a fully discrete approximation of semilinear stochastic PDEs driven by multiplicative Gaussian and Poisson noise

This paper investigates the weak convergence of a fully discrete approximation for a class of semilinear stochastic partial differential equations (SPDEs) driven by multiplicative Gaussian and Poisson noise. The scheme combines a spectral Galerkin discretization in space with an exponential Euler discretization in time. Compared with single-noise settings, the simultaneous presence of continuous and jump perturbations introduces additional challenges arising from discontinuous trajectories, nonlocal jump operators, and their interaction with state-dependent multiplicative coefficients. To overcome these difficulties, we establish the exponential integrability of the numerical solution processes and refined moment estimates in negative Sobolev norms for the associated first- and second-order variational processes. Combining these estimates with a discrete-time approximation argument and Wiener--Poisson Malliavin calculus, we derive regularity estimates for the solutions of the Kolmogorov equations associated with the spatial Galerkin approximations. Based on these estimates, we obtain the temporal and spatial weak convergence orders arbitrarily close to one with respect to discretization parameters $τ$ and $λ_N^{-1}$, respectively. To the best of our knowledge, this provides the first weak convergence result for a fully discrete approximation of semilinear SPDEs with simultaneous multiplicative Gaussian and Poisson noise.

math.NA↗

Elastic Selective Spectral Hybrids for Train-Once, Export-Many Budgeted Inference

Deploying a language model under different computing and latency budgets calls for compact models with different quality-cost trade-offs. Towards this end, elastic spectral state space models provide ordered and temporally decomposed channels that can be truncated, but they are associated with linear time-invariant filters that cannot selectively preserve relevant past information or forget irrelevant information as the context evolves. To address this, we introduce the Elastic Selective Spectral Hybrid (ESSH), which realizes each Hankel spectral channel as an independent recurrent unit using fitted damped rotation modes. It also features an input-dependent decay and write/read gates that make temporal retention and state update input-dependent while preserving channel-wise truncation and a structured recurrence for efficient execution. ESSH combines these selective spectral mixers with sliding-window attention and jointly trains multiple capacities by reducing spectral-channel count and feed-forward width at different rates through a two-rate capacity map with full-model distillation. The resulting models support chunked parallel training and fused recurrent decoding while avoiding computation for discarded channels. At full capacity, ESSH achieves language-modeling quality comparable to similarly sized independently trained models, while smaller exports exhibit a smooth quality-cost trade-off. We validate the effectiveness of the proposed framework using language understanding, retrieval, cross-domain text, and DNA experiments, by assessing quality retention and the trade-off against independently trained and elastic baselines. At the 1.53B model configuration, fused batch-one decoding takes 1.37 ms per token on a B300, providing a 2.14-2.80x speedup over the tested Mamba-2 and Mamba-3 implementations and 3.03x over Transformer++ at matched parameter counts.

cs.LG↗

Random attractors and almost-sure stability under discretization of a stochastic autoparametric system

For a stochastic autoparametric block-and-pendulum system, the long-time dynamics exhibit two fundamental features: the almost-sure stability of the single mode solution, characterized by its Lyapunov exponent, and the global asymptotic dynamics when this single mode solution loses stability. This naturally raises the question of whether these dynamical features are preserved under discretization, since such preservation is essential for the resulting discrete system to faithfully capture the qualitative behavior of the continuous system. To address this question, we first establish the existence of a random attractor for the continuous system subject to multiplicative stochastic excitation, providing a rigorous characterization of the global asymptotic dynamics. We then propose a numerical discretization that induces a discrete random dynamical system and prove the convergence of its random attractor to the continuous one as the step size tends to zero. In addition, we show that the numerical Lyapunov exponent of the single mode solution has the same sign as its continuous counterpart for sufficiently small step sizes, thus preserving the corresponding almost-sure stability or instability classification. These results demonstrate that the proposed discretization captures both the global asymptotic dynamics and the stability characteristics of the underlying stochastic autoparametric system.

math.DS↗

Symplectic methods for stochastic Hamiltonian systems: asymptotic error distributions and Hamiltonian-specific analysis

In this paper, we investigate the asymptotic error distributions of symplectic methods for stochastic Hamiltonian systems and further provide Hamiltonian-specific analysis that clarifies the superiority of symplectic methods. Our contribution is threefold. First, we derive the asymptotic error distributions of symplectic methods for stochastic Hamiltonian systems with multiplicative noise and additive noise, respectively, and show that the obtained limiting stochastic processes satisfy equations retaining the Hamiltonian formulations. Second, we propose a new approach for calculating the asymptotic error distribution, revealing the connection between the stochastic modified equation and the asymptotic error distribution. Third, we characterize the limiting distribution of the normalized Hamiltonian deviation, thereby illustrating through test equations the superiority of symplectic methods for long-time simulations of the Hamiltonians, even in the limit as the step size tends to zero.

math.NA↗

The law of iterated logarithm for numerical approximation of time-homogeneous Markov process

The law of the iterated logarithm (LIL) for the time-homogeneous Markov process with a unique invariant measure characterizes the almost sure maximum possible fluctuation of time averages around the ergodic limit. Whether a numerical approximation can preserve this asymptotic pathwise behavior remains an open problem. In this work, we give a positive answer to this question and establish the LIL for the numerical approximation of such a process under verifiable assumptions. The Markov process is discretized by a decreasing time-step strategy, which yields the non-homogeneous numerical approximation but facilitates a martingale-based analysis. The key ingredient in proving the LIL for such numerical approximation lies in extracting a quasi-uniform time-grid subsequence from the original non-uniform time grids and establishing the LIL for a predominant martingale along it, while the remainder terms converge to zero. Finally, we illustrate that our results can be flexibly applied to numerical approximations of a broad class of stochastic systems, including SODEs and SPDEs.

math.NA↗

sqrtVINS: Robust and Ultrafast Square-Root Filter-based 3D Motion Tracking

In this paper, we develop and open-source, for the first time, a square-root filter (SRF)-based visual-inertial navigation system (VINS), termed sqrtVINS, which is ultra-fast, numerically stable, and capable of dynamic initialization even under extreme conditions (i.e., extremely small time window). Despite recent advancements in VINS, resource constraints and numerical instability on embedded (robotic) systems with limited precision remain critical challenges. A square-root covariance-based filter offers a promising solution by providing numerical stability, efficient memory usage, and guaranteed positive semi-definiteness. However, canonical SRFs suffer from inefficiencies caused by disruptions in the triangular structure of the covariance matrix during updates. The proposed method significantly improves VINS efficiency with a novel Cholesky decomposition (LLT)-based SRF update, by fully exploiting the system structure to preserve the structure. Moreover, we design a fast, robust, dynamic initialization method, which first recovers the minimal states without triangulating 3D features and then efficiently performs iterative SRF update to refine the full states, enabling seamless VINS operation. The proposed LLT-based SRF is extensively verified through numerical studies, demonstrating superior numerical stability and achieving robust efficient performance on 32-bit single-precision floats, operating at twice the speed of state-of-the-art (SOTA) methods. Our initialization method, tested on both mobile workstations and Jetson Nano computers, achieving a high success rate of initialization even within a 100 ms window under minimal conditions. Finally, the proposed sqrtVINS is extensively validated across diverse scenarios, demonstrating strong efficiency, robustness, and reliability. The full open-source implementation is released to support future research and applications.

cs.RO↗

Random attractor and SRB measure for stochastic Hopf bifurcation under discretization

Chaotic phases in stochastic differential equations are characterized by two essential long-time dynamical features: a random attractor capturing asymptotic geometry and a Sinai-Ruelle-Bowen (SRB) measure describing statistical information. This paper investigates whether the stochastic Hopf bifurcation under discretization could inherit both features. We establish that the stochastic Hopf bifurcation under discretization induces a discrete random dynamical system. Further, we prove that this discrete system possesses a random attractor, and then derive the existence of an SRB measure by demonstrating a strictly positive numerical Lyapunov exponent. Numerical experiments visualize the retained random attractor and SRB measure for the discrete random dynamical system, revealing structures consistent with the theoretical chaotic phase.

math.DS↗

$L^p$-strong convergence orders of fully discrete schemes for the SPDE driven by Lévy noise

It is well known that for a stochastic differential equation driven by Lévy noise, the temporal Hölder continuity in $L^p$ sense of the exact solution does not exceed $1/p$. This leads to that the $L^p$-strong convergence order of a numerical scheme will vanish as $p$ increases to infinity if the temporal Hölder continuity of the solution process is directly used. A natural question arises: can one obtain the $L^p$-strong convergence order that does not depend on $p$? In this paper, we provide a positive answer for fully discrete schemes of the stochastic partial differential equation (SPDE) driven by Lévy noise. Two cases are considered: the first is the linear multiplicative Poisson noise with $ν(χ)<\infty$ and the second is the additive Poisson noise with $ν(χ)\leq\infty$, where $ν$ is the Lévy measure and $χ$ is the mark set. For the first case, we present a strategy by employing the jump-adapted time discretization, while for the second case, we introduce the approach based on the recently obtained Lê's quantitative John--Nirenberg inequality. We show that proposed schemes converge in $L^p$ sense with orders almost $1/2$ in both space and time for all $p\ge2$, which contributes novel results in the numerical analysis of the SPDE driven by Lévy noise.

math.NA↗

A new class of splitting methods that preserve ergodicity and exponential integrability for stochastic Langevin equation

In this paper, we propose a new class of splitting methods to solve the stochastic Langevin equation, which can simultaneously preserve the ergodicity and exponential integrability of the original equation. The central idea is to extract a stochastic subsystem that possesses the strict dissipation from the original equation, which is inspired by the inheritance of the Lyapunov structure for obtaining the ergodicity. We prove that the exponential moment of the numerical solution is bounded, thus validating the exponential integrability of the proposed methods. Further, we show that under moderate verifiable conditions, the methods have the first-order convergence in both strong and weak senses, and we present several concrete splitting schemes based on the methods. The splitting strategy of methods can be readily extended to construct conformal symplectic methods and high-order methods that preserve both the ergodicity and the exponential integrability, as demonstrated in numerical experiments. Our numerical experiments also show that the proposed methods have good performance in the long-time simulation.

math.NA↗

On structure preservation for fully discrete finite difference schemes of stochastic heat equation with Lévy space-time white noise

This paper investigates the structure preservation and convergence analysis of a class of fully discrete finite difference schemes for the stochastic heat equation driven by Lévy space-time white noise. The novelty lies in the simultaneous preservation of intrinsic structures for the exact solution, in particular the weak intermittency of moments and the regularity of càdlàg path in negative fractional Sobolev spaces. The key in the proof is the detailed analysis of technical estimates for discrete Green functions of the numerical solution. This analysis is also crucial in establishing the mean-square convergence of the schemes with orders of almost $\frac12$ in space and almost $\frac14$ in time.

math.NA↗

Superiority of stochastic symplectic methods via the law of iterated logarithm

The superiority of stochastic symplectic methods over non-symplectic counterparts has been verified by plenty of numerical experiments, especially in capturing the asymptotic behaviour of the underlying solution process. How can one theoretically explain this superiority? This paper gives an answer to this problem from the perspective of the law of iterated logarithm, taking the linear stochastic Hamiltonian system in Hilbert space as a test model. The main contribution is twofold. First, by fully utilizing the time-change theorem for martingales and the Borell--TIS inequality, we prove that the upper limit of the exact solution with a specific scaling function almost surely equals some non-zero constant, thus confirming the validity of the law of iterated logarithm. Second, we prove that stochastic symplectic fully discrete methods asymptotically preserve the law of iterated logarithm, but non-symplectic ones do not. This reveals the good ability of stochastic symplectic methods in characterizing the almost sure asymptotic growth of the utmost fluctuation of the underlying solution process. Applications of our results to the linear stochastic oscillator and the linear stochastic Schrodinger equation are also presented.

math.NA↗

Longtime behaviors of $θ$-Euler-Maruyama method for stochastic functional differential equations

This paper investigates longtime behaviors of the $θ$-Euler-Maruyama method for the stochastic functional differential equation with superlinearly growing coefficients. We focus on the longtime convergence analysis in mean-square sense and weak sense of the $θ$-Euler-Maruyama method, the convergence of the numerical invariant measure, the existence and convergence of the numerical density function, and the Freidlin-Wentzell large deviation principle of the method. The main contributions are outlined as follows. First, we obtain the longtime mean-square convergence of the $θ$-Euler-Maruyama method and show that the mean-square convergence rate is $\frac12$. A key step in the proof is to establish the time-independent boundedness of high-order moments of the numerical functional solution. Second, based on the technique of the Malliavin calculus, we present the longtime weak convergence of the $θ$-Euler-Maruyama method, which implies that the invariant measure of the $θ$-Euler-Maruyama functional solution converges to the exact one with rate $1.$ Third, by the analysis of the test-functional-independent weak convergence and negative moment estimates of the determinant of the corresponding Malliavin covariance matrix, we derive the existence, convergence, and the logarithmic estimate of the density function of the $θ$-Euler-Maruyama solution. At last, utilizing the weak convergence method, we obtain the Freidlin-Wentzell large deviation principle for the $θ$-Euler-Maruyama solution on the infinite time horizon.

math.NA↗

Long-time weak convergence analysis of a semi-discrete scheme for stochastic Maxwell equations

It is known from the monograph [1, Chapter 5] that the weak convergence analysis of numerical schemes for stochastic Maxwell equations is an unsolved problem. This paper aims to fill the gap by establishing the long-time weak convergence analysis of the semi-implicit Euler scheme for stochastic Maxwell equations. Based on analyzing the regularity of transformed Kolmogorov equation associated to stochastic Maxwell equations and constructing a proper continuous adapted auxiliary process for the semi-implicit scheme, we present the long-time weak convergence analysis for this scheme and prove that the weak convergence order is one, which is twice the strong convergence order. As applications of this result, we obtain the convergence order of the numerical invariant measure, the strong law of large numbers and central limit theorem related to the numerical solution, and the error estimate of the multi-level Monte Carlo estimator. As far as we know, this is the first result on the weak convergence order for stochastic Maxwell equations.

math.NA↗

NeRF-VINS: A Real-time Neural Radiance Field Map-based Visual-Inertial Navigation System

Achieving efficient and consistent localization a prior map remains challenging in robotics. Conventional keyframe-based approaches often suffers from sub-optimal viewpoints due to limited field of view (FOV) and/or constrained motion, thus degrading the localization performance. To address this issue, we design a real-time tightly-coupled Neural Radiance Fields (NeRF)-aided visual-inertial navigation system (VINS). In particular, by effectively leveraging the NeRF's potential to synthesize novel views, the proposed NeRF-VINS overcomes the limitations of traditional keyframe-based maps (with limited views) and optimally fuses IMU, monocular images, and synthetically rendered images within an efficient filter-based framework. This tightly-coupled fusion enables efficient 3D motion tracking with bounded errors. We extensively compare the proposed NeRF-VINS against the state-of-the-art methods that use prior map information and demonstrate its ability to perform real-time localization, at over 10 Hz, on a resource-constrained Jetson AGX Orin embedded platform.

cs.RO↗

Long-time dynamics of stochastic wave equation with dissipative damping and its full discretization: exponential ergodicity and strong law of large numbers

For stochastic wave equation, when the dissipative damping is a non-globally Lipschitz function of the velocity, there are few results on the long-time dynamics, in particular, the exponential ergodicity and strong law of large numbers, for the equation and its numerical discretization to our knowledge. Focus on this issue, the main contributions of this paper are as follows. First, based on constructing novel Lyapunov functionals, we show the unique invariant measure and exponential ergodicity of the underlying equation and its full discretization. Second, the error estimates of invariant measures both in Wasserstein distance and in the weak sense are obtained. Third, the strong laws of large numbers of the equation and the full discretization are obtained, which states that the time averages of the exact and numerical solutions are shown to converge to the ergodic limit almost surely.

math.PR↗

Strong convergence of adaptive time-stepping schemes for the stochastic Allen--Cahn equation

It is known in \cite{beccari} that the standard explicit Euler-type scheme (such as the exponential Euler and the linear-implicit Euler schemes) with a uniform timestep, though computationally efficient, may diverge for the stochastic Allen--Cahn equation. To overcome the divergence, this paper proposes and analyzes adaptive time-stepping schemes, which adapt the timestep at each iteration to control numerical solutions from instability. The \textit{a priori} estimates in $\mathcal {C}(\mathcal {O})$-norm and $\dot{H}^β(\mathcal{O})$-norm of numerical solutions are established provided the adaptive timestep function is suitably bounded, which plays a key role in the convergence analysis. We show that the adaptive time-stepping schemes converge strongly with order $\fracβ{2}$ in time and $\fracβ{d}$ in space with $d$ ($d=1,2,3$) being the dimension and $β\in(0,2]$. Numerical experiments show that the adaptive time-stepping schemes are simple to implement and at a lower computational cost than a scheme with the uniform timestep.

math.NA↗