NSTL国家科技图书文献中心

1. CONTROLLING STOCHASTIC GRADIENT DESCENT USING STOCHASTIC APPROXIMATION FOR ROBUST DISTRIBUTED OPTIMIZATION NSTL国家科技图书文献中心

Adit Jain | Vikram Krishnamurthy - 《Numerical Algebra, Control and Optimization》 - 2025,15(1) - 173～195 - 共23页

摘要： stochastic approximation algorithm to search for the | controlling the stochastic* gradient descent, performed by* | querying a stochastic* oracle. Each query has a learning* | optimal policy using a sigmoidal approximation* and does* | This paper deals with the problem of

关键词： Stochastic approximation | Distributed optimization | Markov decision processes

2. Online Learning Under a Separable Stochastic Approximation Framework NSTL国家科技图书文献中心

Min Gan | Xiang-xiang Su ... - 《IEEE Transactions on Pattern Analysis and Machine Intelligence》 - 2025,47(2) - 1317～1330 - 共14页

摘要： separable stochastic* approximation framework. The central* | viewed as a stochastic* approximation variant of block* | (RLS) algorithm, akin to a stochastic* Newton method* | nonlinear parameters are adjusted using the stochastic | We propose an online learning algorithm

关键词： Stochastic processes | Approximation algorithms | Optimization | Machine learning algorithms | Machine learning | Convergence | Artificial neural networks | Training | Minimization | Convex functions

3. Distributed Population-Based Simultaneous Perturbation Stochastic Approximation for Fine-Tuning Large Language Models NSTL国家科技图书文献中心

Yajing Tan | Yuwei Huang ... - 《PRICAI 2024,Part III》 - Pacific Rim International Conference on Artificial Intelligence - 2025, - 21～26 - 共6页

摘要： (called MeZO) of simultaneous perturbation stochastic | approximation (SPSA), one well-established zeroth-order | Very recently, a memory-efficient version | optimizer from the automatic control community, has shown | competitive fine-tuning performance on large language models

关键词： Population-based evolution | Simultaneous perturbation stochastic approximation | Fine-Tuning of large language models

4. CVaR stochastic programming model for monotone stochastic tensor complementarity problem by using its penalized sample average approximation algorithm NSTL国家科技图书文献中心

Xu, Yuncheng | Liu, Sanyang ... - 《Journal of Computational and Applied Mathematics》 - 2025,460 - 共12页

摘要： stochastic tensor complementarity problem, where the | expectation of the involved stochastic* tensor is a strictly* | at risk stochastic* programming (CVaR-SP) model of* | monotone stochastic* tensor complementarity problem (STCP* | stochastic residual defined by the modified restricted NCP

关键词： Stochastic tensor complementarity problem | Conditional Value-at-Risk | Strictly positive semi-definite tensor | Penalized sample average approximation algorithm

5. Adaptive Deep Density Approximation for Stochastic Dynamical Systems NSTL国家科技图书文献中心

Junjie He | Qifeng Liao ... - 《Journal of scientific computing》 - 2025,102(3) - 1.1～1.30 - 共30页

摘要： deep neural network approximation for stochastic | associated with the stochastic* dynamical systems, a new* | Abstract In this paper we consider adaptive | dynamical systems. Based on the continuity equation | temporal KRnet (tKRnet) is proposed to approximate the

关键词： 34F05 | 60H35 | 62M45 | 65C30

6. Stochastic transport with Levy noise fully discrete numerical approximation NSTL国家科技图书文献中心

Andreas Stein | Andrea Barth - 《Mathematics and computers in simulation》 - 2025,227(Jan.) - 347～370 - 共24页

摘要：Semilinear hyperbolic stochastic* partial* | discrete approximation* scheme that tackles these* | Galerkin scheme for spatial approximation, derived | Karhunen-Loeve expansions. This approximation* yields a* | differential equations (SPDEs) find widespread applications

关键词： Numerical analysis of SPDEs | Stochastic transport equation | Infinite-dimensional Levy processes | Discontinuous Galerkin method

7. Distributed System Identification for Linear Stochastic Systems Under an Adaptive Event‐Triggered Scheme NSTL国家科技图书文献中心

Xiaoxue Geng | Wenxiao Zhao - 《International journal of adaptive control and signal processing》 - 2025,39(3) - 471～488 - 共18页

摘要： stochastic approximation algorithm with expanding | identification problem for linear stochastic* systems whose* | ABSTRACT This article considers a distributed | input and output observations are scheduled by an | adaptive eventtriggered scheme. An event detector with

关键词： distributed identification | eventtriggered scheme | stochastic approximation | strong consistency

8. Stochastic Halfspace Approximation Method for Convex Optimization With Nonsmooth Functional Constraints NSTL国家科技图书文献中心

Nitesh Kumar Singh | Ion Necoara - 《IEEE Transactions on Automatic Control》 - 2025,70(1) - 479～486 - 共8页

摘要： stochastic halfspace approximation and we provide a unified | functional constraints. We propose a new stochastic | gradient algorithm, called the stochastic* halfspace* | approximation method (SHAM), to solve this problem, where at | onto one halfspace approximation* of a randomly chosen*

关键词： Convergence | Convex functions | Linear programming | Approximation algorithms | Signal processing algorithms | Optimization | Systems engineering and theory

9. An Euler-Maruyama method for Caputo-Hadamard fractional stochastic differential equations on exponential meshes and its fast approximation NSTL国家科技图书文献中心

Li, Min | Guo, Shangjiang ... - 《Numerical algorithms》 - 2025,99(2) - 573～601 - 共29页

摘要： Caputo-Hadamard fractional stochastic* differential* | approximation to decrease the computational cost of the EM | This paper studies the numerical solutions of | equations. Firstly, we construct an Euler-Maruyama (EM | ) scheme for the equations, and the corresponding

关键词： Caputo-Hadamard fractional stochastic differential equations | EM scheme | Sum-of-exponentials approximation | Fast EM scheme | Multilevel Monte Carlo simulation | Strong convergence

10. Nonnegativity preserving convolution kernels. Application to Stochastic Volterra Equations in closed convex domains and their approximation NSTL国家科技图书文献中心

Alfonsi, Aurelien - 《Stochastic Processes and Their Applications》 - 2025,181 - 共23页

摘要： convolution kernel like in Stochastic* Volterra Equations or* | analyze the stochastic* invariance of a closed convex set* | by Stochastic* Volterra Equations. We also get a* | approximation scheme for the weak error that stays in the | This work defines and studies one-dimensional

摘要： stochastic approximation algorithm to search for the | controlling the stochastic* gradient descent, performed by* | querying a stochastic* oracle. Each query has a learning* | optimal policy using a sigmoidal approximation* and does* | This paper deals with the problem of

关键词： Stochastic approximation | Distributed optimization | Markov decision processes

摘要： separable stochastic* approximation framework. The central* | viewed as a stochastic* approximation variant of block* | (RLS) algorithm, akin to a stochastic* Newton method* | nonlinear parameters are adjusted using the stochastic | We propose an online learning algorithm

关键词： Stochastic processes | Approximation algorithms | Optimization | Machine learning algorithms | Machine learning | Convergence | Artificial neural networks | Training | Minimization | Convex functions

摘要： (called MeZO) of simultaneous perturbation stochastic | approximation (SPSA), one well-established zeroth-order | Very recently, a memory-efficient version | optimizer from the automatic control community, has shown | competitive fine-tuning performance on large language models

关键词： Population-based evolution | Simultaneous perturbation stochastic approximation | Fine-Tuning of large language models

关键词： Stochastic tensor complementarity problem | Conditional Value-at-Risk | Strictly positive semi-definite tensor | Penalized sample average approximation algorithm

摘要： deep neural network approximation for stochastic | associated with the stochastic* dynamical systems, a new* | Abstract In this paper we consider adaptive | dynamical systems. Based on the continuity equation | temporal KRnet (tKRnet) is proposed to approximate the

关键词： 34F05 | 60H35 | 62M45 | 65C30

摘要：Semilinear hyperbolic stochastic* partial* | discrete approximation* scheme that tackles these* | Galerkin scheme for spatial approximation, derived | Karhunen-Loeve expansions. This approximation* yields a* | differential equations (SPDEs) find widespread applications

关键词： Numerical analysis of SPDEs | Stochastic transport equation | Infinite-dimensional Levy processes | Discontinuous Galerkin method

摘要： stochastic approximation algorithm with expanding | identification problem for linear stochastic* systems whose* | ABSTRACT This article considers a distributed | input and output observations are scheduled by an | adaptive eventtriggered scheme. An event detector with

关键词： distributed identification | eventtriggered scheme | stochastic approximation | strong consistency

摘要： stochastic halfspace approximation and we provide a unified | functional constraints. We propose a new stochastic | gradient algorithm, called the stochastic* halfspace* | approximation method (SHAM), to solve this problem, where at | onto one halfspace approximation* of a randomly chosen*

关键词： Convergence | Convex functions | Linear programming | Approximation algorithms | Signal processing algorithms | Optimization | Systems engineering and theory

摘要： Caputo-Hadamard fractional stochastic* differential* | approximation to decrease the computational cost of the EM | This paper studies the numerical solutions of | equations. Firstly, we construct an Euler-Maruyama (EM | ) scheme for the equations, and the corresponding

关键词： Caputo-Hadamard fractional stochastic differential equations | EM scheme | Sum-of-exponentials approximation | Fast EM scheme | Multilevel Monte Carlo simulation | Strong convergence

摘要： convolution kernel like in Stochastic* Volterra Equations or* | analyze the stochastic* invariance of a closed convex set* | by Stochastic* Volterra Equations. We also get a* | approximation scheme for the weak error that stays in the | This work defines and studies one-dimensional

关键词： Completely monotone kernels | Stochastic Volterra Equations | Volterra Equations | Hawkes processes with inhibition | Rough Heston model

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： stochastic approximation algorithm to search for the | controlling the stochastic gradient descent, performed by | querying a stochastic oracle. Each query has a learning | optimal policy using a sigmoidal approximation and does | This paper deals with the problem of

关键词： Stochastic approximation | Distributed optimization | Markov decision processes

摘要： separable stochastic approximation framework. The central | viewed as a stochastic approximation variant of block | (RLS) algorithm, akin to a stochastic Newton method | nonlinear parameters are adjusted using the stochastic | We propose an online learning algorithm

关键词： Stochastic processes | Approximation algorithms | Optimization | Machine learning algorithms | Machine learning | Convergence | Artificial neural networks | Training | Minimization | Convex functions

摘要： (called MeZO) of simultaneous perturbation stochastic | approximation (SPSA), one well-established zeroth-order | Very recently, a memory-efficient version | optimizer from the automatic control community, has shown | competitive fine-tuning performance on large language models

关键词： Population-based evolution | Simultaneous perturbation stochastic approximation | Fine-Tuning of large language models

摘要： stochastic tensor complementarity problem, where the | expectation of the involved stochastic tensor is a strictly | at risk stochastic programming (CVaR-SP) model of | monotone stochastic tensor complementarity problem (STCP | stochastic residual defined by the modified restricted NCP

关键词： Stochastic tensor complementarity problem | Conditional Value-at-Risk | Strictly positive semi-definite tensor | Penalized sample average approximation algorithm

摘要： deep neural network approximation for stochastic | associated with the stochastic dynamical systems, a new | Abstract In this paper we consider adaptive | dynamical systems. Based on the continuity equation | temporal KRnet (tKRnet) is proposed to approximate the

关键词： 34F05 | 60H35 | 62M45 | 65C30

摘要：Semilinear hyperbolic stochastic partial | discrete approximation scheme that tackles these | Galerkin scheme for spatial approximation, derived | Karhunen-Loeve expansions. This approximation yields a | differential equations (SPDEs) find widespread applications

关键词： Numerical analysis of SPDEs | Stochastic transport equation | Infinite-dimensional Levy processes | Discontinuous Galerkin method

摘要： stochastic approximation algorithm with expanding | identification problem for linear stochastic systems whose | ABSTRACT This article considers a distributed | input and output observations are scheduled by an | adaptive eventtriggered scheme. An event detector with

关键词： distributed identification | eventtriggered scheme | stochastic approximation | strong consistency

摘要： stochastic halfspace approximation and we provide a unified | functional constraints. We propose a new stochastic | gradient algorithm, called the stochastic halfspace | approximation method (SHAM), to solve this problem, where at | onto one halfspace approximation of a randomly chosen

关键词： Convergence | Convex functions | Linear programming | Approximation algorithms | Signal processing algorithms | Optimization | Systems engineering and theory

摘要： Caputo-Hadamard fractional stochastic differential | approximation to decrease the computational cost of the EM | This paper studies the numerical solutions of | equations. Firstly, we construct an Euler-Maruyama (EM | ) scheme for the equations, and the corresponding

关键词： Caputo-Hadamard fractional stochastic differential equations | EM scheme | Sum-of-exponentials approximation | Fast EM scheme | Multilevel Monte Carlo simulation | Strong convergence

摘要： convolution kernel like in Stochastic Volterra Equations or | analyze the stochastic invariance of a closed convex set | by Stochastic Volterra Equations. We also get a | approximation scheme for the weak error that stays in the | This work defines and studies one-dimensional

关键词： Completely monotone kernels | Stochastic Volterra Equations | Volterra Equations | Hawkes processes with inhibition | Rough Heston model

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： stochastic approximation algorithm to search for the | controlling the stochastic* gradient descent, performed by* | querying a stochastic* oracle. Each query has a learning* | optimal policy using a sigmoidal approximation* and does* | This paper deals with the problem of

摘要： separable stochastic* approximation framework. The central* | viewed as a stochastic* approximation variant of block* | (RLS) algorithm, akin to a stochastic* Newton method* | nonlinear parameters are adjusted using the stochastic | We propose an online learning algorithm

摘要： deep neural network approximation for stochastic | associated with the stochastic* dynamical systems, a new* | Abstract In this paper we consider adaptive | dynamical systems. Based on the continuity equation | temporal KRnet (tKRnet) is proposed to approximate the

摘要：Semilinear hyperbolic stochastic* partial* | discrete approximation* scheme that tackles these* | Galerkin scheme for spatial approximation, derived | Karhunen-Loeve expansions. This approximation* yields a* | differential equations (SPDEs) find widespread applications

摘要： stochastic approximation algorithm with expanding | identification problem for linear stochastic* systems whose* | ABSTRACT This article considers a distributed | input and output observations are scheduled by an | adaptive eventtriggered scheme. An event detector with

摘要： Caputo-Hadamard fractional stochastic* differential* | approximation to decrease the computational cost of the EM | This paper studies the numerical solutions of | equations. Firstly, we construct an Euler-Maruyama (EM | ) scheme for the equations, and the corresponding

摘要： convolution kernel like in Stochastic* Volterra Equations or* | analyze the stochastic* invariance of a closed convex set* | by Stochastic* Volterra Equations. We also get a* | approximation scheme for the weak error that stays in the | This work defines and studies one-dimensional