paper-with-me

Papers

Object-Oriented Dynamics Learning through Multi-Level Abstraction

2019-04-16 · Guangxiang Zhu, Jianhao Wang, Zhizhou Ren, Zichuan Lin, Chongjie Zhang

Object-based approaches for learning action-conditioned dynamics has demonstrated promise for generalization and interpretability. However, existing approaches suffer from structural limitations and optimization difficulties for common environments with multiple dynamic objects. In this paper, we present a novel self-supervised learning framework, called Multi-level Abstraction Object-oriented Predictor (MAOP), which employs a three-level learning architecture that enables efficient object-based dynamics learning from raw visual observations. We also design a spatial-temporal relational reasoning mechanism for MAOP to support instance-level dynamics learning and handle partial observability. Our results show that MAOP significantly outperforms previous methods in terms of sample efficiency and generalization over novel environments for learning environment models. We also demonstrate that learned dynamics models enable efficient planning in unseen environments, comparable to true environment models. In addition, MAOP learns semantically and visually interpretable disentangled representations.

📄 PDF Abstract BibTeX arXiv:1904.07482

Code (1)

mig-zh/OODP tf

Tasks

ObjectRelational ReasoningSelf-Supervised Learning

Similar Papers 제목 키워드 기반

Object-Oriented Model Learning through Multi-Level Abstraction

2019-05-01 · ICLR 2019 5 · Guangxiang Zhu, Jianhao Wang, Zhizhou Ren, Chongjie Zhang

Object-based approaches for learning action-conditioned dynamics has demonstrated promise for generalization and interpretability. However, existing approaches suffer from structural limitations and optimization difficu…

modelObjectRelational ReasoningSelf-Supervised Learning

Object-Oriented Dynamics Predictor

2018-05-25 · NeurIPS 2018 12 · Guangxiang Zhu, Zhiao Huang, Chongjie Zhang

Generalization has been one of the major challenges for learning dynamics models in model-based reinforcement learning. However, previous work on action-conditioned dynamics prediction focuses on learning the pixel-level…

Model-based Reinforcement LearningObjectReinforcement Learning

Learning Causal Dynamics Models in Object-Oriented Environments

2024-05-21 · Zhongwei Yu, Jingqing Ruan, Dengpeng Xing

Causal dynamics models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. To learn CDMs, recent studies have performed causal discovery to capture the causal depend…

Causal DiscoveryComputational EfficiencyObject

TONAV: Task-Oriented Navigation and Action-Velocity Chunk Learning for Articulated Object Quadrupedal Mobile Manipulation

2026-08-23 · Haoran Lin, Mingyu Yang, Pengfei Qi, Kehan Chen 외 arxiv

Quadruped mobile manipulation requires two tightly coupled capabilities: reaching manipulation-ready configurations and maintaining stable contact throughout articulated-object interaction. However, existing methods ofte…

SoftGPT: Learn Goal-oriented Soft Object Manipulation Skills by Generative Pre-trained Heterogeneous Graph Transformer

2023-06-22 · Junjia Liu, Zhihao LI, WanYu Lin, Sylvain Calinon 외

Soft object manipulation tasks in domestic scenes pose a significant challenge for existing robotic skill learning techniques due to their complex dynamics and variable shape characteristics. Since learning new manipulat…

Object