mlsanityData to 14 Oct 2026

Towards a Grounded Theory of Causation for Embodied AI

arXiv 2206.13973 ↗First posted 28 Jun 2022cs.AI cs.LG stat.MLcited by 9

Not on the map. Why?

Abstract (arXiv)

There exist well-developed frameworks for causal modelling, but these require rather a lot of human domain expertise to define causal variables and perform interventions. In order to enable autonomous agents to learn abstract causal models through interactive experience, the existing theoretical foundations need to be extended and clarified. Existing frameworks give no guidance regarding variable choice / representation, and more importantly, give no indication as to which behaviour policies or physical transformations of state space shall count as interventions. The framework sketched in this paper describes actions as transformations of state space, for instance induced by an agent running a policy. This makes it possible to describe in a uniform way both transformations of the micro-state space and abstract models thereof, and say when the latter is veridical / grounded / natural. We then introduce (causal) variables, define a mechanism as an invariant predictor, and say when an action can be viewed as a ``surgical intervention'', thus bringing the objective of causal representation & intervention skill learning into clearer focus.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 28 of its 54 references that are in the map. First posted on arXiv 28 Jun 2022.

Builds on

28 of 54 references.

References · 28

  1. 2013Auto-Encoding Variational Bayescited by 4,053
  2. 2019Invariant Risk Minimizationcited by 1,854
  3. 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,167
  4. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 941
  5. 2020Object-Centric Learning with Slot Attentioncited by 876
  6. 2018Diversity is All You Need: Learning Skills without a Reward Functioncited by 752
  7. 2015Causal inference using invariant prediction: identification and confidence intervalscited by 536
  8. 2021Causal Abstractions of Neural Networkscited by 422
  9. 2021Towards Causal Representation Learningcited by 209
  10. 2020Representation Learning via Invariant Causal Mechanismscited by 199
  11. 2022Weakly supervised causal representation learningcited by 140
  12. 2022CITRIS: Causal Identifiability from Temporal Intervened Sequencescited by 133
16 more
  1. 2017Causal Consistency of Structural Equation Modelscited by 118
  2. 2018Abstracting Causal Modelscited by 107
  3. 2021Desiderata for Representation Learning: A Causal Perspectivecited by 76
  4. 2019Approximate Causal Abstractioncited by 63
  5. 2020Natural Graph Networkscited by 35
  6. 2016From Deterministic ODEs to Dynamic Structural Causal Modelscited by 32
  7. 2020Emergent Real-World Robotic Skills via Unsupervised Off-Policy Reinforcement Learningcited by 31
  8. 2015Multi-Level Cause-Effect Systemscited by 30
  9. 2022A Theory of Abstraction in Reinforcement Learningcited by 27
  10. 2018Causal Modeling of Dynamical Systemscited by 24
  11. 2013Cyclic Causal Discovery from Continuous Equilibrium Datacited by 21
  12. 2020Amortized learning of neural causal representationscited by 14
  13. 2017Independently Controllable Factorscited by 12
  14. 2017Independently Controllable Featurescited by 11
  15. 2022Learning by Doing: Controlling a Dynamical System using Causality, Control, and Reinforcement Learningcited by 7
  16. 2021Causality and independence in perfectly adapted dynamical systemscited by 4

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 4; 2025: 2.

9 papers.

  1. 2024Position: Categorical Deep Learning is an Algebraic Theory of All Architecturescited by 32
  2. 2023Causal Triplet: An Open Challenge for Intervention-centric Causal Representation Learningcited by 22
  3. 2023Causal models in string diagramscited by 14
  4. 2024Identifiable Causal Representation Learning: Unsupervised, Multi-View, and Multi-Environmentcited by 12
  5. 2023A Measure-Theoretic Axiomatisation of Causalitycited by 8
  6. 2025What is causal about causal models and representations?cited by 6
  7. 2024Systems with Switching Causal Relations: A Meta-Causal Perspectivecited by 2
  8. 2025Causality in the human niche: lessons for machine learningcited by 1
  9. 2024Generative AI Systems: A Systems-based Perspective on Generative AIcited by 0