Model Based Reinforcement Learning

Model-based reinforcement learning (MBRL) aims to improve the sample efficiency and robustness of reinforcement learning agents by learning a model of the environment's dynamics. Current research focuses on enhancing model accuracy and robustness through techniques like incorporating expert knowledge, using bisimulation metrics for state representation, and employing adversarial training to handle uncertainties. These advancements are leading to more efficient and reliable control policies in various applications, including robotics, autonomous driving, and even protein design, by reducing the need for extensive real-world interactions during training.

Papers

October 3, 2023

Probabilistic Reach-Avoid for Bayesian Neural Networks
Matthew Wicker, Luca Laurenti, Andrea Patane, Nicola Paoletti, Alessandro Abate, Marta Kwiatkowska
Optimal Policy Bayesian Neural Network Model Based Reinforcement Learning Control Synthesis Reach Avoid

October 1, 2023

Uncertainty-aware hybrid paradigm of nonlinear MPC and model-based RL for offroad navigation: Exploration of transformers in the predictive model
Faraz Lotfi, Khalil Virji, Farnoosh Faraji, Lucas Berry, Andrew Holliday, David Meger, Gregory Dudek
Model Based Reinforcement Learning Predictive Model Uncertainty Aware Road Navigation Autonomous Driving Model Nonlinear MPC

September 30, 2023

HarmonyDream: Task Harmonization Inside World Models
Haoyu Ma, Jialong Wu, Ningya Feng, Chenjun Xiao, Dong Li, Jianye Hao, Jianmin Wang, Mingsheng Long
Model Based Reinforcement Learning Reward Model World Model Inter Vendor Harmonization

September 26, 2023

Towards High Efficient Long-horizon Planning with Expert-guided Motion-encoding Tree Search
Tong Zhou, Erli Lyu, Jiaole Wang, Guangdu Cen, Ziqi Zha, Senmao Qi, Max Q. -H. Meng
Autonomous Driving Model Based Reinforcement Learning Target Search

September 25, 2023

MoDem-V2: Visuo-Motor World Models for Real-World Robot Manipulation
Patrick Lancaster, Nicklas Hansen, Aravind Rajeswaran, Vikash Kumar
Real World Model Based Reinforcement Learning World Model Vision Based Agent Exploration Visual Model Based

September 16, 2023

DOMAIN: MilDly COnservative Model-BAsed OfflINe Reinforcement Learning
Xiao-Yin Liu, Xiao-Hu Zhou, Xiao-Liang Xie, Shi-Qi Liu, Zhen-Qiu Feng, Hao Li, Mei-Jiang Gui, Tian-Yu Xiang, De-Xing Huang, Zeng-Guang Hou
Reinforcement Learning Model Based Reinforcement Learning Model Uncertainty Internet Service Domain Model Based Offline Reinforcement Learning

September 11, 2023

Mind the Uncertainty: Risk-Aware and Actively Exploring Model-Based Reinforcement Learning
Marin Vlastelica, Sebastian Blaes, Cristina Pineri, Georg Martius
High Uncertainty Anticipation Model Based Reinforcement Learning Safety Constraint Safety Critical Control Risk Management Trajectory Sampling

August 31, 2023

August 10, 2023

Exploring the Potential of World Models for Anomaly Detection in Autonomous Driving
Daniel Bogdoll, Lukas Bosch, Tim Joseph, Helen Gremmelmaier, Yitian Yang, J. Marius Zöllner
Autonomous Driving Anomaly Detection Autonomous Vehicle Full Potential Model Based Reinforcement Learning World Model

August 4, 2023

Learning to Shape by Grinding: Cutting-surface-aware Model-based Reinforcement Learning
Takumi Hachimine, Jun Morimoto, Takamitsu Matsubara
LeArning Abstract Model Based Reinforcement Learning Object Shape Hitting Time Cutting Plane

July 24, 2023

Theoretically Guaranteed Policy Improvement Distilled from Model-Based Planning
Chuming Li, Ruonan Jia, Jie Liu, Yinmin Zhang, Yazhe Niu, Yaodong Yang, Yu Liu, Wanli Ouyang
Model Based Reinforcement Learning Policy Improvement Model Based Planning Control Benchmark

June 30, 2023

$\lambda$-models: Effective Decision-Aware Reinforcement Learning with Latent Models
Claas A Voelcker, Arash Ahmadian, Romina Abachi, Igor Gilitschenski, Amir-massoud Farahmand
Model Based Reinforcement Learning Latent Variable Model Lambda Calculus Decision Loss Decision Aware

June 27, 2023

Trajectory Generation, Control, and Safety with Denoising Diffusion Probabilistic Models
Nicolò Botteghi, Federico Califano, Mannes Poel, Christoph Brune
External Control Model Based Reinforcement Learning Optimal Control Control Barrier Function Denoising Diffusion Probabilistic Model Human SAFETY Trajectory Generation Safe Trajectory

June 15, 2023

Deep Generative Models for Decision-Making and Control
Michael Janner
External Control Decision Making Model Based Reinforcement Learning Trajectory Optimization Model Based Generative Tool

June 6, 2023

Model-Based Reinforcement Learning with Multi-Task Offline Pretraining
Minting Pan, Yitao Zheng, Yunbo Wang, Xiaokang Yang
Reinforcement Learning Model Based Reinforcement Learning World Model Transferable Representation Behaviour Learning

June 1, 2023

May 29, 2023

Pre-training Contextualized World Models with In-the-wild Videos for Reinforcement Learning
Jialong Wu, Haoyu Ma, Chaoyi Deng, Mingsheng Long
Reinforcement Learning Model Based Reinforcement Learning World Model Unsupervised Pre Training Wild Image Video Pretraining

May 26, 2023

Adaptive PD Control using Deep Reinforcement Learning for Local-Remote Teleoperation with Stochastic Time Delays
Luc McCutcheon, Saber Fallah
Reinforcement Learning Deep Reinforcement Learning Model Based Reinforcement Learning Robot Teleoperation Delay Differential Equation Stochastic Delay Remote Manipulation

Model Based Reinforcement Learning

Papers

Probabilistic Reach-Avoid for Bayesian Neural Networks

Uncertainty-aware hybrid paradigm of nonlinear MPC and model-based RL for offroad navigation: Exploration of transformers in the predictive model

HarmonyDream: Task Harmonization Inside World Models

Towards High Efficient Long-horizon Planning with Expert-guided Motion-encoding Tree Search

MoDem-V2: Visuo-Motor World Models for Real-World Robot Manipulation

DOMAIN: MilDly COnservative Model-BAsed OfflINe Reinforcement Learning

Mind the Uncertainty: Risk-Aware and Actively Exploring Model-Based Reinforcement Learning

RePo: Resilient Model-Based Reinforcement Learning by Regularizing Posterior Predictability

The Power of MEME: Adversarial Malware Creation with Model-Based Reinforcement Learning

Exploring the Potential of World Models for Anomaly Detection in Autonomous Driving

Learning to Shape by Grinding: Cutting-surface-aware Model-based Reinforcement Learning

Theoretically Guaranteed Policy Improvement Distilled from Model-Based Planning

$\lambda$-models: Effective Decision-Aware Reinforcement Learning with Latent Models

Trajectory Generation, Control, and Safety with Denoising Diffusion Probabilistic Models

Deep Generative Models for Decision-Making and Control

Model-Based Reinforcement Learning with Multi-Task Offline Pretraining

IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Control

What model does MuZero learn?

Pre-training Contextualized World Models with In-the-wild Videos for Reinforcement Learning

Adaptive PD Control using Deep Reinforcement Learning for Local-Remote Teleoperation with Stochastic Time Delays