世界模型 - arXivDaily 专题

2608.01208 2026-08-04 q-fin.MF cs.LG q-fin.RM 新提交 80%

Climate-Dyna Deep Hedging for XVAs: Model-Based Reinforcement Learning, Residual Climate HVA, and Hedge-Instrument Discovery

面向XVAs的Climate-Dyna深度对冲：基于模型的强化学习、剩余气候HVA及对冲工具发现

Xiaozhen Wang, Francois Buet-Golfouse

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

AI总结该研究提出Climate-Dyna深度对冲方法，通过基于模型的强化学习解决剩余气候HVA问题，在EU ETS半合成研究中有效降低了气候费用，显著减少了遗憾值并保留了大部分精确辅助增益。

Comments 15 pages, 2 figures, 1 table

2512.08108 2025-12-10 cs.LG cs.AI 80%

Scalable Offline Model-Based RL with Action Chunks

可扩展的基于模型的强化学习与动作块

Kwanyoung Park, Seohong Park, Youngwoon Lee, Sergey Levine

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

AI总结本文提出MAC方法，通过动作块模型和拒绝采样提升离线基于模型的强化学习在复杂长 horizon 任务中的性能。

Comments 22 pages, 7 figures

2110.02758 2023-02-21 cs.LG cs.AI cs.RO 80%

Mismatched No More: Joint Model-Policy Optimization for Model-Based RL

Benjamin Eysenbach, Alexander Khazatsky, Sergey Levine, Ruslan Salakhutdinov

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

Comments NeurIPS 2022

2206.04551 2022-06-10 cs.LG cs.AI cs.RO 80%

A Relational Intervention Approach for Unsupervised Dynamics Generalization in Model-Based Reinforcement Learning

Jixian Guo, Mingming Gong, Dacheng Tao

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

Comments ICLR2022 accepted paper

2006.16712 2022-04-01 cs.LG cs.AI stat.ML 80%

Model-based Reinforcement Learning: A Survey

Thomas M. Moerland, Joost Broekens, Aske Plaat, Catholijn M. Jonker

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

2106.13229 2021-08-10 cs.LG cs.AI cs.RO 80%

Model-Based Reinforcement Learning via Latent-Space Collocation

Oleh Rybkin, Chuning Zhu, Anusha Nagabandi, Kostas Daniilidis, Igor Mordatch, Sergey Levine

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

Comments International Conference on Machine Learning (ICML), 2021. Videos and code at https://orybkin.github.io/latco/

1907.02057 2019-07-04 cs.LG cs.AI cs.RO stat.ML 80%

Benchmarking Model-Based Reinforcement Learning

Tingwu Wang, Xuchan Bao, Ignasi Clavera, Jerrick Hoang, Yeming Wen, Eric Langlois, Shunshi Zhang, Guodong Zhang, Pieter Abbeel, Jimmy Ba

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

Comments 8 main pages, 8 figures; 14 appendix pages, 25 figures

2512.04279 2025-12-30 cs.RO 80%

Driving Beyond Privilege: Distilling Dense-Reward Knowledge into Sparse-Reward Policies

超越特权：将密集奖励知识蒸馏到稀疏奖励策略中

Feeza Khan Khanzada, Jaerock Kwon

机构 * Department of Electrical and Computer Engineering, University of Michigan-Dearborn（电气与计算机工程系，密歇根大学迪尔伯恩分校）

专题命中模型式强化学习：world model(abstract)；world model(abstract)；latent dynamics(abstract)；model-based reinforcement learning(abstract)

AI总结本文提出通过密集奖励蒸馏学习稀疏奖励策略，提升自动驾驶在稀疏目标上的泛化能力。

2601.21306 2026-05-25 cs.LG cs.AI 79%

The Surprising Difficulty of Search in Model-Based Reinforcement Learning

基于模型的强化学习中搜索的惊人困难

Wei-Di Chang, Mikael Henaff, Brandon Amos, Gregory Dudek, Scott Fujimoto

机构 * Meta FAIR ； McGill University（麦吉尔大学）

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

AI总结本文研究基于模型的强化学习中的搜索问题，发现即使模型高度准确，搜索也可能损害性能，而缓解过估计偏差比提高模型或价值函数精度更重要，通过取价值函数集成的最小值可有效解决偏差并实现有效搜索。

Comments ICML 2026

2510.05957 2025-10-08 cs.RO 79%

Learning to Crawl: Latent Model-Based Reinforcement Learning for Soft Robotic Adaptive Locomotion

Vaughn Gzenda, Robin Chhabra

机构 * Embodied Learning and Intelligence for eXploration and Innovative soft Robotics (ELIXIR) Lab（具身学习与智能探索与创新软机器人实验室）

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

2408.14685 2025-07-28 physics.flu-dyn cs.LG 79%

Model-Based Reinforcement Learning for Control of Strongly-Disturbed Unsteady Aerodynamic Flows

Zhecheng Liu, Diederik Beckers, Jeff D. Eldredge

机构 * Department of Mechanical and Aerospace Engineering（机械与航空航天工程系）； Graduate Aerospace Laboratories（航空航天研究生实验室）

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

Journal ref AIAA Journal, (Early Access) 2025

2504.04164 2025-07-04 cs.LG 79%

MInCo: Mitigating Information Conflicts in Distracted Visual Model-based Reinforcement Learning

Shiguang Sun, Hanbo Zhang, Zeyang Liu, Xinrui Yang, Lipeng Wan, Xingyu Chen, Xuguang Lan

机构 * National Key Laboratory of Human-Machine Hybrid Augmented Intelligence（人机混合增强智能国家重点实验室）； National Engineering Research Center for Visual Information and Applications（视觉信息与应用国家工程研究中心）； Institute of Artificial Intelligence and Robotics（人工智能与机器人研究院）； Xi’an Jiaotong University（西安交通大学）； National University of Singapore（新加坡国立大学）

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

2503.08728 2025-03-13 cs.MA cs.AI 79%

Enhancing Traffic Signal Control through Model-based Reinforcement Learning and Policy Reuse

Yihong Li, Chengwei Zhang, Furui Zhan, Wanting Liu, Kailing Zhou, Longji Zheng

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

2405.16184 2024-05-28 eess.SY cs.AI cs.LG cs.SY 79%

Safe Deep Model-Based Reinforcement Learning with Lyapunov Functions

Harry Zhang

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

2404.09946 2024-04-16 cs.LG cs.AI stat.ML 79%

A Note on Loss Functions and Error Compounding in Model-based Reinforcement Learning

Nan Jiang

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

2302.01244 2023-02-03 cs.LG 79%

Is Model Ensemble Necessary? Model-based RL via a Single Model with Lipschitz Regularized Value Function

Ruijie Zheng, Xiyao Wang, Huazhe Xu, Furong Huang

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

Comments ICLR 2023

2209.05530 2022-09-14 cs.LG cs.AI 79%

Model-based Reinforcement Learning with Multi-step Plan Value Estimation

Haoxin Lin, Yihao Sun, Jiaji Zhang, Yang Yu

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

2204.11464 2022-06-28 cs.LG cs.AI 79%

Towards Evaluating Adaptivity of Model-Based Reinforcement Learning Methods

Yi Wan, Ali Rahimi-Kalahroudi, Janarthanan Rajendran, Ida Momennejad, Sarath Chandar, Harm van Seijen

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

2106.14080 2022-01-31 cs.LG cs.AI stat.ML 79%

Model-Advantage and Value-Aware Models for Model-Based Reinforcement Learning: Bridging the Gap in Theory and Practice

Nirbhay Modhe, Harish Kamath, Dhruv Batra, Ashwin Kalyan

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

2010.10740 2020-10-22 cs.LG cs.RO cs.SY eess.SY 79%

Safety Verification of Model Based Reinforcement Learning Controllers

Akshita Gupta, Inseok Hwang

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

1905.11527 2019-11-01 cs.LG cs.AI stat.ML 79%

Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies

Yonathan Efroni, Nadav Merlis, Mohammad Ghavamzadeh, Shie Mannor

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

Comments NeurIPS 2019

1808.09105 2019-06-25 cs.LG cs.RO stat.ML 79%

SOLAR: Deep Structured Representations for Model-Based Reinforcement Learning

Marvin Zhang, Sharad Vikram, Laura Smith, Pieter Abbeel, Matthew J. Johnson, Sergey Levine

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

Comments ICML 2019. Project website: https://sites.google.com/view/icml19solar

1805.09496 2019-06-06 cs.LG cs.AI stat.ML 79%

Intelligent Trainer for Model-Based Reinforcement Learning

Yuanlong Li, Linsen Dong, Xin Zhou, Yonggang Wen, Kyle Guan

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

Comments 13 pages

1812.07671 2019-01-30 cs.LG cs.AI cs.RO stat.ML 79%

Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RL

Anusha Nagabandi, Chelsea Finn, Sergey Levine

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

Comments Project website: https://sites.google.com/berkeley.edu/onlineviameta

2502.14819 2025-10-30 cs.LG 79%

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

Vlad Sobal, Wancong Zhang, Kyunghyun Cho, Randall Balestriero, Tim G. J. Rudner, Yann LeCun

机构 * New York University（纽约大学）； Genentech（基因泰克）； Brown University（布朗大学）； University of Toronto（多伦多大学）； Meta – FAIR

专题命中模型式强化学习：latent dynamics(title,abstract)；dynamics model(title,abstract)；分类 cs.LG

Comments Project web page: https://latent-planning.github.io/

2005.06800 2020-06-30 cs.LG stat.ML 79%

Context-aware Dynamics Model for Generalization in Model-Based Reinforcement Learning

Kimin Lee, Younggyo Seo, Seunghyun Lee, Honglak Lee, Jinwoo Shin

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

Comments Accepted in ICML2020. First two authors contributed equally, website: https://sites.google.com/view/cadm code: https://github.com/younggyoseo/CaDM

2608.18946 2026-08-25 quant-ph cs.AI 版本更新 78%

AlphaClifford: Efficient Clifford Synthesis and Transpilation with Model-based RL

AlphaClifford：基于模型强化学习的高效Clifford电路综合与 transpilation

Daniele Lizzio Bosco, Jacopo Cossio, Carla Piazza, Giuseppe Serra

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

AI总结本研究提出基于模型强化学习的AlphaClifford框架，用于高效综合Clifford电路，在总门数和CNOT门数上优于现有方法，还可用于硬件约束 transpilation及Clifford+T流程优化，为量子编译提供可扩展途径。

Comments Accepted manuscript to appear in the proceedings of the IEEE International Conference on Quantum Computing and Engineering (QCE 2026). 11 pages, 3 figures. v2: Added discussion and citation of concurrent work by Yeung, Kissinger, and Cornish (arXiv:2605.10910)

1610.03518 2026-06-04 cs.RO cs.AI cs.LG cs.SY eess.SY 78%

Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Model

通过学习深度逆动力学模型实现仿真到现实世界的迁移

Paul Christiano, Zain Shah, Igor Mordatch, Jonas Schneider, Trevor Blackwell, Joshua Tobin, Pieter Abbeel, Wojciech Zaremba

专题命中模型式强化学习：world model(abstract)；dynamics model(title,abstract)；world model(abstract)；分类 cs.AI、cs.LG、cs.RO

AI总结本文提出通过学习深度逆动力学模型，在仿真与现实世界之间实现控制策略的迁移，解决仿真与现实差异导致的性能下降问题。

2509.00215 2025-09-05 cs.RO cs.AI cs.LG 78%

First Order Model-Based RL through Decoupled Backpropagation

Joseph Amigo, Rooholla Khorrambakht, Elliot Chane-Sane, Nicolas Mansard, Ludovic Righetti

机构 * Machines in Motion Laboratory, New York University, USA（纽约大学运动机器实验室）； LAAS-CNRS, Université de Toulouse, CNRS, Toulouse, France（图卢兹大学LAAS-CNRS）； Artificial and Natural Intelligence Toulouse Institute, Toulouse, France（图卢兹人工智能与自然智能研究所）

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

Comments CoRL 2025. Project website: https://machines-in-motion.github.io/DMO/

2505.13144 2025-05-20 cs.LG cs.AI cs.RO 78%

Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning

Dongsu Lee, Minhae Kwon

机构 * Carnegie Mellon University, Pittsburgh, USA（卡内基梅隆大学）； Department of Intelligent Semiconductors, Soongsil University, Seoul, South Korea（智能半导体系，顺斯大学）

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

Comments 2025 ICML

1. 模型式强化学习 1126 篇

Climate-Dyna Deep Hedging for XVAs: Model-Based Reinforcement Learning, Residual Climate HVA, and Hedge-Instrument Discovery

Scalable Offline Model-Based RL with Action Chunks

Mismatched No More: Joint Model-Policy Optimization for Model-Based RL

A Relational Intervention Approach for Unsupervised Dynamics Generalization in Model-Based Reinforcement Learning

Model-based Reinforcement Learning: A Survey

Model-Based Reinforcement Learning via Latent-Space Collocation

Benchmarking Model-Based Reinforcement Learning

Driving Beyond Privilege: Distilling Dense-Reward Knowledge into Sparse-Reward Policies

The Surprising Difficulty of Search in Model-Based Reinforcement Learning

Learning to Crawl: Latent Model-Based Reinforcement Learning for Soft Robotic Adaptive Locomotion

Model-Based Reinforcement Learning for Control of Strongly-Disturbed Unsteady Aerodynamic Flows

MInCo: Mitigating Information Conflicts in Distracted Visual Model-based Reinforcement Learning

Enhancing Traffic Signal Control through Model-based Reinforcement Learning and Policy Reuse

Safe Deep Model-Based Reinforcement Learning with Lyapunov Functions

A Note on Loss Functions and Error Compounding in Model-based Reinforcement Learning

Is Model Ensemble Necessary? Model-based RL via a Single Model with Lipschitz Regularized Value Function

Model-based Reinforcement Learning with Multi-step Plan Value Estimation

Towards Evaluating Adaptivity of Model-Based Reinforcement Learning Methods

Model-Advantage and Value-Aware Models for Model-Based Reinforcement Learning: Bridging the Gap in Theory and Practice

Safety Verification of Model Based Reinforcement Learning Controllers

Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies

SOLAR: Deep Structured Representations for Model-Based Reinforcement Learning

Intelligent Trainer for Model-Based Reinforcement Learning

Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RL

Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

Context-aware Dynamics Model for Generalization in Model-Based Reinforcement Learning

AlphaClifford: Efficient Clifford Synthesis and Transpilation with Model-based RL

Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Model

First Order Model-Based RL through Decoupled Backpropagation

Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning