mlsanityData to 7 Oct 2026

Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics

arXiv 1706.04317 ↗First posted 14 Jun 2017cs.AIcited by 57

Not on the map. Why?

Abstract (arXiv)

The recent adaptation of deep neural network-based methods to reinforcement learning and planning domains has yielded remarkable progress on individual tasks. Nonetheless, progress on task-to-task transfer remains limited. In pursuit of efficient and robust generalization, we introduce the Schema Network, an object-oriented generative physics simulator capable of disentangling multiple causes of events and reasoning backward through causes to achieve goals. The richly structured architecture of the Schema Network can learn the dynamics of an environment directly from data. We compare Schema Networks with Asynchronous Advantage Actor-Critic and Progressive Networks on a suite of Breakout variations, reporting results on training efficiency and zero-shot generalization, consistently demonstrating faster, more robust learning and better transfer. We argue that generalizing from limited data and learning causal relationships are essential abilities on the path toward generally intelligent systems.

Key facts. Cited by 57 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Jun 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 1; 2020: 0; 2021: 15; 2022: 7; 2023: 15; 2024: 7; 2025: 6; 2026: 5.

57 papers.

Sort:

By area:

  1. 2021A Survey of Zero-shot Generalisation in Deep Reinforcement Learningcited by 164
  2. 2024WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environmentcited by 99
  3. 2021A Survey on Interpretable Reinforcement Learningcited by 70
  4. 2023An Investigation into Pre-Training Object-Centric Representations for Reinforcement Learningcited by 49
  5. 2025What Has a Foundation Model Found? Using Inductive Bias to Probe for World Modelscited by 46
  6. 2021Abstract Spatial-Temporal Reasoning via Probabilistic Abduction and Executioncited by 43
  7. 2021Learning Neuro-Symbolic Relational Transition Models for Bilevel Planningcited by 43
  8. 2021Human-Level Reinforcement Learning through Theory-Based Modeling, Exploration, and Planningcited by 43
  9. 2023A Survey on Causal Reinforcement Learningcited by 43
  10. 2019Few-Shot Bayesian Imitation Learning with Logical Program Policiescited by 30
  11. 2021Learning Algebraic Representation for Systematic Generalization in Abstract Reasoningcited by 27
  12. 2022Symmetry-Based Representations for Artificial and Biological General Intelligencecited by 26
38 more
  1. 2025SIMA 2: A Generalist Embodied Agent for Virtual Worldscited by 26
  2. 2025PoE-World: Compositional World Modeling with Products of Programmatic Expertscited by 25
  3. 2023Causal Reinforcement Learning: A Surveycited by 22
  4. 2021Capturing the objects of vision with neural networkscited by 21
  5. 2023Explore to Generalize in Zero-Shot RLcited by 20
  6. 2021HALMA: Humanlike Abstraction Learning Meets Affordance in Rapid Problem Solvingcited by 15
  7. 2022On the link between conscious function and general intelligence in humans and machinescited by 15
  8. 2022FAIR for AI: An interdisciplinary and international community building perspectivecited by 13
  9. 2025Synthesizing world models for bilevel planningcited by 13
  10. 2021Discovering User-Interpretable Capabilities of Black-Box Planning Agentscited by 10
  11. 2021Theoretically Principled Deep RL Acceleration via Nearest Neighbor Function Approximationcited by 10
  12. 2018LIFT: Reinforcement Learning in Computer Systems by Learning From Demonstrationscited by 8
  13. 2023Conceptual Reinforcement Learning for Language-Conditioned Taskscited by 8
  14. 2023PDSketch: Integrated Planning Domain Programming and Learningcited by 7
  15. 2024From Frege to chatGPT: Compositionality in language, cognition, and deep neural networkscited by 7
  16. 2026OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3cited by 7
  17. 2022Object-Category Aware Reinforcement Learningcited by 6
  18. 2023Granger Causal Interaction Skill Chainscited by 5
  19. 2023PushWorld: A benchmark for manipulation planning with tools and movable obstaclescited by 3
  20. 2023DetermiNet: A Large-Scale Diagnostic Dataset for Complex Visually-Grounded Referencing using Determinerscited by 3
  21. 2026World Models: A Comprehensive Survey of Architectures, Methodologies, Reasoning Paradigms, and Applicationscited by 3
  22. 2021Spatially Invariant Unsupervised 3D Object-Centric Learning and Scene Decompositioncited by 2
  23. 2023Q-Cogni: An Integrated Causal Reinforcement Learning Frameworkcited by 2
  24. 2023A Review of Symbolic, Subsymbolic and Hybrid Methods for Sequential Decision Makingcited by 2
  25. 2023Procedural Content Generation via Knowledge Transformation (PCG-KT)cited by 2
  26. 2026SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systemscited by 2
  27. 2021A Classification of Artificial Intelligence Systems for Mathematics Educationcited by 1
  28. 2023The Construction of Reality in an AI: A Reviewcited by 1
  29. 2023The Role of Diverse Replay for Generalisation in Reinforcement Learningcited by 1
  30. 2023Efficient Open-world Reinforcement Learning via Knowledge Distillation and Autonomous Rule Discoverycited by 1
  31. 2024Learning to Play Video Games with Intuitive Physics Priorscited by 1
  32. 2024Training on Irrelevant States Implies Data Augmentation: Generalization in Contextual MDPscited by 1
  33. 2025How Ensembles of Distilled Policies Improve Generalisation in Reinforcement Learningcited by 1
  34. 2025Causality in the human niche: lessons for machine learningcited by 1
  35. 2021Probabilistic Programming Bots in Intuitive Physics Game Playcited by 0
  36. 2021Plug and Play, Model-Based Reinforcement Learningcited by 0
  37. 2021Feature-Attending Recurrent Modules for Generalization in Reinforcement Learningcited by 0
  38. 2022Does DQN really learn? Exploring adversarial training schemes in Pongcited by 0

50 most cited of 57 All 57 (JSON)