mlsanityData to 14 Oct 2026

Entity-Centric Reinforcement Learning for Object Manipulation from Pixels

arXiv 2404.01220 ↗First posted 1 Apr 2024cs.RO cs.CV cs.LGcited by 27

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

Manipulating objects is a hallmark of human intelligence, and an important task in domains such as robotics. In principle, Reinforcement Learning (RL) offers a general approach to learn object manipulation. In practice, however, domains with more than a few objects are difficult for RL agents due to the curse of dimensionality, especially when learning from raw image observations. In this work we propose a structured approach for visual RL that is suitable for representing multiple objects and their interaction, and use it to learn goal-conditioned manipulation of several objects. Key to our method is the ability to handle goals with dependencies between the objects (e.g., moving objects in a certain order). We further relate our architecture to the generalization capability of the trained agent, based on a theoretical result for compositional generalization, and demonstrate agents that learn with 3 objects but generalize to similar tasks with over 10 objects. Videos and code are available on the project website: https://sites.google.com/view/entity-centric-rl

Key facts. Cited by 27 papers in the map (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 48 of its 57 references that are in the map. First posted on arXiv 1 Apr 2024.

Builds on

48 of 57 references.

References · 48

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2017Proximal Policy Optimization Algorithmscited by 15,215
  4. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  5. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  6. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  7. 2015Spatial Transformer Networkscited by 1,429
  8. 2017Deep Setscited by 1,421
  9. 2018Relational inductive biases, deep learning, and graph networkscited by 1,360
  10. 2023Mastering Diverse Domains through World Modelscited by 1,253
  11. 2017Hindsight Experience Replaycited by 1,136
  12. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
36 more
  1. 2020Object-Centric Learning with Slot Attentioncited by 876
  2. 2015End-to-End Training of Deep Visuomotor Policiescited by 875
  3. 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 743
  4. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 360
  5. 2017A simple neural network module for relational reasoningcited by 346
  6. 2019Improving Sample Efficiency in Model-Free Reinforcement Learning from Imagescited by 274
  7. 2018Visual Reinforcement Learning with Imagined Goalscited by 251
  8. 2018Graph networks as learnable physics engines for inference and controlcited by 244
  9. 2019Skew-Fit: State-Covering Self-Supervised Reinforcement Learningcited by 174
  10. 2015Deep Spatial Autoencoders for Visuomotor Learningcited by 153
  11. 2021Illiterate DALL-E Learns to Composecited by 147
  12. 2022SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Modelscited by 147
  13. 2021Discovering and Achieving Goals via World Modelscited by 131
  14. 2018Unsupervised Learning of Object Landmarks through Conditional Image Generationcited by 114
  15. 2021Density-aware Chamfer Distance as a Comprehensive Metric for Point Cloud Completioncited by 112
  16. 2019SCALOR: Generative World Models with Scalable Object Representationscited by 108
  17. 2022Learning Multi-Object Dynamics with Compositional Neural Radiance Fieldscited by 63
  18. 2019COBRA: Data-Efficient Model-Based RL through Unsupervised Object Discovery and Curiosity-Driven Explorationcited by 57
  19. 2019Towards Practical Multi-Object Manipulation using Relational Reinforcement Learningcited by 55
  20. 2023An Investigation into Pre-Training Object-Centric Representations for Reinforcement Learningcited by 49
  21. 2020Self-supervised Visual Reinforcement Learning with Object-centric Representationscited by 44
  22. 2019Structured Object-Aware Physics Prediction for Video Modeling and Planningcited by 43
  23. 2022Curious Exploration via Structured World Models Yields Zero-Shot Object Manipulationcited by 31
  24. 2022Policy Architectures for Compositional Generalization in Controlcited by 23
  25. 2022Toward Compositional Generalization in Object-Oriented World Modelingcited by 22
  26. 2022Visuomotor Control in Multi-Object Scenes Using Object-Aware Representationscited by 21
  27. 2022Learning What and Where: Disentangling Location and Identity Tracking Without Supervisioncited by 21
  28. 2022Compositional Multi-Object Reinforcement Learning with Linear Relation Networkscited by 19
  29. 2022Unsupervised Image Representation Learning with Deep Latent Particlescited by 19
  30. 2023DDLP: Unsupervised Object-Centric Video Prediction with Deep Dynamic Latent Particlescited by 19
  31. 2023FOCUS: Object-Centric World Models for Robotics Manipulationcited by 17
  32. 2021Self-supervised Reinforcement Learning with Independently Controllable Subgoalscited by 16
  33. 2023Learning Dynamic Attribute-factored World Models for Efficient Multi-object Reinforcement Learningcited by 15
  34. 2023A Survey on Compositional Generalization in Applicationscited by 12
  35. 2023Efficient RL via Disentangled Environment and Agent Representationscited by 12
  36. 2022Object-based active inferencecited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 11; 2026: 10.

27 papers.

Sort:

By area:

  1. 2024Temporally Consistent Object-Centric Learning by Contrasting Slotscited by 31
  2. 2024Zero-Shot Object-Centric Representation Learningcited by 19
  3. 2025Object-Centric World Model for Language-Guided Manipulationcited by 16
  4. 2026Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modelingcited by 15
  5. 2025CTRL-O: Language-Controllable Object-Centric Visual Representation Learningcited by 12
  6. 2024Entity-based Reinforcement Learning for Autonomous Cyber Defencecited by 11
  7. 2025MetaSlot: Break Through the Fixed Number of Slots in Object-Centric Learningcited by 11
  8. 2025Learning Interactive World Model for Object-Centric Reinforcement Learningcited by 11
  9. 2025Object-Centric World Models from Few-Shot Annotations for Sample-Efficient Reinforcement Learningcited by 10
  10. 2024EC-Diffuser: Multi-Object Manipulation via Entity-Centric Behavior Generationcited by 8
  11. 2026Hierarchical Entity-centric Reinforcement Learning with Factored Subgoal Diffusioncited by 7
  12. 2025Object-Centric Pretraining via Target Encoder Bootstrappingcited by 6
15 more
  1. 2025Object-Centric Representations Improve Policy Generalization in Robot Manipulationcited by 6
  2. 2025Focusing on What Matters: Object-Agent-centric Tokenization for Vision Language Action modelscited by 6
  3. 2025Is an object-centric representation beneficial for robotic manipulation ?cited by 3
  4. 2026Improved Object-Centric Diffusion Learning with Registers and Contrastive Alignmentcited by 3
  5. 2024Efficient Exploration and Discriminative World Model Learning with an Object-Centric Abstractioncited by 2
  6. 2024Object-centric proto-symbolic behavioural reasoning from pixelscited by 2
  7. 2025Human-Allied Relational Reinforcement Learningcited by 1
  8. 20263D-DLP: Self-Supervised 3D Object-Centric Scene Representation Learningcited by 1
  9. 2025Task and Skill Planning: Hierarchical Robot Planning with Black-Box Skillscited by 0
  10. 2026ConceptACT: Episode-Level Concepts for Sample-Efficient Robotic Imitation Learningcited by 0
  11. 2026Are Object-Centric Representations Better At Compositional Generalization?cited by 0
  12. 2026GRAIL: Autonomous Concept Grounding for Neuro-Symbolic Reinforcement Learningcited by 0
  13. 2026Learning Bilevel Policies over Symbolic World Models for Long-Horizon Planningcited by 0
  14. 2026Learning in Low-Dimensional Subspaces: Orthogonal Bottlenecks for Reinforcement Learningcited by 0
  15. 2026ParticleSplat: Self-supervised Object-centric Latent Particle Splattingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Learning Category-Level Generalizable Object Manipulation Policy via Generative Adversarial Self-Imitation Learning from Demonstrations
  2. 2025Generalizable Hierarchical Skill Learning via Object-Centric Representation
  3. 2025ObjectVLA: End-to-End Open-World Object Manipulation Without Demonstration
  4. 2025FLEX: A Framework for Learning Robot-Agnostic Force-based Skills Involving Sustained Contact Object Manipulation
  5. 2025Versatile and Generalizable Manipulation via Goal-Conditioned Reinforcement Learning with Grounded Object Detection
  6. 2025ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models
  7. 2024SOLD: Slot Object-Centric Latent Dynamics Models for Relational Manipulation Learning from Pixels
  8. 2025Zero-Shot Visual Generalization in Robot Manipulation