世界模型

Adaptive Model-Based Reinforcement Learning for Orbit Feedback Control in NSLS-II Storage Ring

自适应模型驱动强化学习用于NSLS-II存储环轨道反馈控制

Zeyu Dong, Yuke Tian, Yu Sun

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

AI总结 本文提出基于模型驱动强化学习的自适应框架，用于NSLS-II存储环轨道反馈控制，通过轨迹优化和在线模型优化实现束流稳定与对准误差最小化。

Comments Accepted by the 20th International Conference on Accelerator and Large Experimental Physics Control Systems (ICALEPCS 2025)

Safe model-based Reinforcement Learning via Model Predictive Control and Control Barrier Functions

通过模型预测控制和控制屏障函数实现安全的模型基于强化学习

Kerim Dzhumageldyev, Filippo Airaldi, Azita Dabiri

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

AI总结 本文提出一种安全的模型基于强化学习框架，结合模型预测控制与控制屏障函数，通过参数化方法改进安全控制策略。

Comments Submitted to IFAC WC 2026, 7 pages, 3 figures

Improved Greedy Identification of Latent Dynamics with Application to Fluid Flows

R. Ayoub, M. Oulghelou, P. J Schmid

专题命中 模型式强化学习 ：latent dynamics(title,abstract)

The Edge-of-Reach Problem in Offline Model-Based Reinforcement Learning

Anya Sims, Cong Lu, Jakob Foerster, Yee Whye Teh

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG；dynamics model(abstract)

Comments Code open-sourced at: https://github.com/anyasims/edge-of-reach

Journal ref NeurIPS 2024

A Safe and Data-efficient Model-based Reinforcement Learning System for HVAC Control

Xianzhong Ding, Zhiyu An, Arya Rathee, Wan Du

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

Learning the Latent dynamics of Fluid flows from High-Fidelity Numerical Simulations using Parsimonious Diffusion Maps

Alessandro Della Pia, Dimitris Patsatzis, Lucia Russo, Constantinos Siettos

专题命中 模型式强化学习 ：latent dynamics(title,abstract)

Journal ref Physics of Fluids 36, 105187 (2024)

Exploiting Symmetry in Dynamics for Model-Based Reinforcement Learning with Asymmetric Rewards

Yasin Sonmez, Neelay Junnarkar, Murat Arcak

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG、cs.RO

FLD: Fourier Latent Dynamics for Structured Motion Representation and Learning

Chenhao Li, Elijah Stanger-Jones, Steve Heim, Sangbae Kim

专题命中 模型式强化学习 ：latent dynamics(title)；分类 cs.AI、cs.LG、cs.RO

Data-Driven Characterization of Latent Dynamics on Quantum Testbeds

Sohail Reddy, Stefanie Guenther, Yujin Cho

专题命中 模型式强化学习 ：latent dynamics(title,abstract)

Live in the Moment: Learning Dynamics Model Adapted to Evolving Policy

Xiyao Wang, Wichayaporn Wongkamjan, Furong Huang

专题命中 模型式强化学习 ：dynamics model(title,abstract)；model-based reinforcement learning(abstract)；model-based RL(abstract)；分类 cs.LG

Comments 16 pages, 5 figures

Journal ref ICML 2023

Model Based Reinforcement Learning with Non-Gaussian Environment Dynamics and its Application to Portfolio Optimization

Huifang Huang, Ting Gao, Pengbo Li, Jin Guo, Peng Zhang, Nan Du

专题命中 模型式强化学习 ：model based reinforcement learning(title)；model-based reinforcement learning(abstract)

Comments arXiv admin note: text overlap with arXiv:2205.15056

Reduced models with nonlinear approximations of latent dynamics for model premixed flame problems

Wayne Isaac Tan Uy, Christopher R. Wentland, Cheng Huang, Benjamin Peherstorfer

专题命中 模型式强化学习 ：latent dynamics(title,abstract)

A Model-Based Reinforcement Learning Approach for PID Design

Hozefa Jesawada, Amol Yerudkar, Carmen Del Vecchio, Navdeep Singh

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

Safe Controller for Output Feedback Linear Systems using Model-Based Reinforcement Learning

S M Nahid Mahmud, Moad Abudia, Scott A Nivison, Zachary I. Bell, Rushikesh Kamalapurkar

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

Comments arXiv admin note: substantial text overlap with arXiv:2110.00271

Model-Based Reinforcement Learning Framework of Online Network Resource Allocation

Bahador Bakhshi, Josep Mangues-Bafalluy

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；model-based RL(abstract)

Comments This is the version of the paper submitted to ICC 2022

Safe Model-Based Reinforcement Learning for Systems with Parametric Uncertainties

S M Nahid Mahmud, Scott A Nivison, Zachary I. Bell, Rushikesh Kamalapurkar

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

Comments This manuscript has been accepted in Frontiers in Robotics and AI. doi: 10.3389/frobt.2021.733104

Safety aware model-based reinforcement learning for optimal control of a class of output-feedback nonlinear systems

S M Nahid Mahmud, Moad Abudia, Scott A Nivison, Zachary I. Bell, Rushikesh Kamalapurkar

专题命中 模型式强化学习 ：model-based reinforcement learning(title,abstract)

Comments arXiv admin note: substantial text overlap with arXiv:2007.12666

Adaptive Optics control using Model-Based Reinforcement Learning

Jalo Nousiainen, Chang Rajani, Markus Kasper, Tapio Helin

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；model-based RL(abstract)

Comments Accepted for publication in Optics Express

Constrained Model-based Reinforcement Learning with Robust Cross-Entropy Method

Zuxin Liu, Hongyi Zhou, Baiming Chen, Sicheng Zhong, Martial Hebert, Ding Zhao

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG、cs.RO

Comments 8 pages, 5 figures

Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning

Lucas Manuelli, Yunzhu Li, Pete Florence, Russ Tedrake

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.RO；predictive model(abstract)；predictive models(abstract)

Physics Consistency and Latent Dynamics in Spatiotemporal Physics Field Generation

基于混合Mamba-Transformer的时空场生成

Peimian Du, Jiabin Liu, Xiaowei Jin, Wangmeng Zuo, Hui Li

专题命中 模型式强化学习 ：latent dynamics(title)；分类 cs.AI、cs.LG

AI总结 本文提出基于混合Mamba-Transformer架构的时空物理场生成模型，并通过物理信息微调机制有效减少物理误差。

SJEPA: Learning Elegant Latent Dynamics with Hybrid Symbolic-Neural Predictors

SJEPA：通过混合符号-神经预测器学习简洁的潜在动力学

Yongchao Huang

专题命中 模型式强化学习 ：latent dynamics(title)；分类 cs.AI、cs.LG

AI总结 SJEPA是一种无重构的JEPA框架，结合符号定律与神经修正学习简洁潜在动力学，在受控摆实验中展现更优性能，揭示了多目标间的可控权衡。

Comments 42 pages

SINDy-RL: Interpretable and Efficient Model-Based Reinforcement Learning

SINDy-RL：可解释且高效的基于模型的强化学习

Nicholas Zolman, Christian Lagemann, Urban Fasel, J. Nathan Kutz, Steven L. Brunton

机构 * Department of Mechanical Engineering, University of Washington, Seattle, WA 98195, USA（华盛顿大学机械工程系） ； Data Science and Artificial Intelligence Department, The Aerospace Corporation, El Segundo, CA 90245（航空航天公司数据科学与人工智能部） ； Department of Aeronautics, Imperial College, London SW7 2AZ, United Kingdom（帝国理工学院航空系） ； Department of Applied Mathematics, University of Washington, Seattle, WA 98195（华盛顿大学应用数学系） ； Department of Electrical and Computer Engineering, University of Washington, Seattle, WA 98195（华盛顿大学电气与计算机工程系） 纠错

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.LG；dynamics model(abstract)

AI总结 本文提出SINDy-RL框架，结合SINDy和DRL，实现低数据下高效、可解释的动力学模型和控制策略，通过基准环境和流体控制实验验证其有效性。

Comments For code, see https://github.com/nzolman/sindy-rl. v2 Update: Included Pinball and 3D Airfoil examples. Christian Lagemann added as an author for contributions with the 3D Airfoil code. To appear in Nature Communications

Journal ref Nat. Commun. 16, 10714 (2025)

Safe Model-based Reinforcement Learning with Stability Guarantees

具有稳定性保证的安全模型基于强化学习

Felix Berkenkamp, Matteo Turchetta, Angela P. Schoellig, Andreas Krause

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG

AI总结 本文提出一种考虑安全性的强化学习算法，通过Lyapunov稳定性验证理论，利用动态统计模型获得具有证明稳定性的高性能控制策略，并在模拟倒立摆中展示其安全优化神经网络策略的能力。

Comments Proc. of Neural Information Processing Systems (NIPS), 2017

Agnostic System Identification for Model-Based Reinforcement Learning

基于模型的强化学习的不可知系统辨识

Stephane Ross, J. Andrew Bagnell

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG

AI总结 针对模型类可能不包含真实系统的不可知情况，提出一种利用无遗憾在线学习算法获得近优策略的迭代方法，并在离散和连续域上验证其有效性。

Comments 8 pages, published in ICML 2012

Geometric Neural Operators via Lie Group-Constrained Latent Dynamics

通过李群约束的潜在动态实现几何神经算子

Jiaquan Zhang, Fachrina Dewi Puspitasari, Songbo Zhang, Yibei Liu, Kuien Liu, Caiyan Qin, Fan Mo, Peng Wang, Yang Yang, Chaoning Zhang

机构 * School of Information and Software Engineering, University of Electronic Science and Technology of China（信息与软件工程学院，电子科学与技术大学） ； Computer Science and Engineering, University of Electronic Science and Technology of China（计算机科学与工程，电子科学与技术大学） ； Institute of Software Chinese Academy of Sciences, Beijing, China（软件研究所，中国科学院） ； School of Robotics and Advanced Manufacture, Harbin Institute of Technology, Shenzhen, China（机器人与先进制造学院，哈尔滨工业大学（深圳）） ； Department of Computer Science, University of Oxford, Oxford, United Kingdom（计算机科学系，牛津大学） 纠错

专题命中 模型式强化学习 ：latent dynamics(title)；分类 cs.AI、cs.LG

AI总结 本文提出了一种基于李群约束的潜在动态方法，用于改进神经算子的几何诱导偏差，从而提高长期预测的保真度。

Dual Alignment Maximin Optimization for Offline Model-based RL

Chi Zhou, Wang Luo, Haoran Li, Congying Han, Tiande Guo, Zicheng Zhang

机构 * UCAS（中国科学院大学） 纠错

专题命中 模型式强化学习 ：model-based RL(title)；分类 cs.AI、cs.LG

Revealing the Challenges of Sim-to-Real Transfer in Model-Based Reinforcement Learning via Latent Space Modeling

Zhilin Lin, Shiliang Sun

机构 * School of Automation and Intelligent Sensing（自动化与智能感知学院） 纠错

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG

Efficient Model-Based Reinforcement Learning Through Optimistic Thompson Sampling

Jasmine Bayrooti, Carl Henrik Ek, Amanda Prorok

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.LG、cs.RO

Comments Appearing in ICLR, 2025

Value-Distributional Model-Based Reinforcement Learning

Carlos E. Luis, Alessandro G. Bottero, Julia Vinogradska, Felix Berkenkamp, Jan Peters

专题命中 模型式强化学习 ：model-based reinforcement learning(title)；分类 cs.AI、cs.LG