mlsanityData to 14 Oct 2026

Sequential Causal Imitation Learning with Unobserved Confounders

arXiv 2208.06276 ↗First posted 12 Aug 2022cs.LG cs.AIcited by 25

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

"Monkey see monkey do" is an age-old adage, referring to na"ive imitation without a deep understanding of a system's underlying mechanics. Indeed, if a demonstrator has access to information unavailable to the imitator (monkey), such as a different set of sensors, then no matter how perfectly the imitator models its perceived environment (See), attempting to reproduce the demonstrator's behavior (Do) can lead to poor outcomes. Imitation learning in the presence of a mismatch between demonstrator and imitator has been studied in the literature under the rubric of causal imitation learning (Zhang et al., 2020), but existing solutions are limited to single-stage decision-making. This paper investigates the problem of causal imitation learning in sequential settings, where the imitator must make multiple decisions per episode. We develop a graphical criterion that is necessary and sufficient for determining the feasibility of causal imitation, providing conditions when an imitator can match a demonstrator's performance despite differing capabilities. Finally, we provide an efficient algorithm for determining imitability and corroborate our theory with simulations.

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 6 of its 28 references that are in the map. First posted on arXiv 12 Aug 2022.

Builds on

6 of 28 references.

Closest ancestors · 3

Same direction, newest first

  1. 2020Causal Transfer for Imitation Learning and Decision Making under Sensor-shift2 Mar
  2. 2019Causal Confusion in Imitation Learning28 May
  3. 2018An Algorithmic Perspective on Imitation Learning16 Nov

Other references · 3

  1. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  2. 2018The highD Dataset: A Drone Dataset of Naturalistic Vehicle Trajectories on German Highways for Validation of Highly Automated Driving Systemscited by 277
  3. 2013Probabilistic Evaluation of Sequential Plans from Causal Models with Hidden Variablescited by 24

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 6; 2024: 7; 2025: 5; 2026: 5.

25 papers.

Sort:

By area:

  1. 2023A Survey on Causal Reinforcement Learningcited by 45
  2. 2023Causal Reinforcement Learning: A Surveycited by 23
  3. 2024Neural Causal Abstractionscited by 21
  4. 2023TGRL: An Algorithm for Teacher Guided Reinforcement Learningcited by 17
  5. 2024Fine-Grained Causal Dynamics Learning with Quantization for Improving Robustness in Reinforcement Learningcited by 11
  6. 2022Instrumental Variables in Causal Inference and Machine Learning: A Surveycited by 10
  7. 2023Delphic Offline Reinforcement Learning under Nonidentifiable Hidden Confoundingcited by 7
  8. 2025Confounding Robust Deep Reinforcement Learning: A Causal Approachcited by 7
  9. 2024Deep Autoregressive Models as Causal Inference Enginescited by 6
  10. 2025Automatic Reward Shaping from Confounded Offline Datacited by 6
  11. 2022Deconfounding Imitation Learning with Variational Inferencecited by 5
  12. 2024Causality from Bottom to Top: A Surveycited by 5
13 more
  1. 2023Initial State Interventions for Deconfounded Imitation Learningcited by 4
  2. 2026Confounding Robust Continuous Control via Automatic Reward Shapingcited by 2
  3. 2023Active and Passive Causal Inference Learningcited by 1
  4. 2024Exploring and Addressing Reward Confusion in Offline Preference Learningcited by 1
  5. 2025Causal Imitation Learning under Expert-Observable and Expert-Unobservable Confoundingcited by 1
  6. 2025On Transportability for Structural Causal Banditscited by 1
  7. 2024DIGIC: Domain Generalizable Imitation Learning by Causal Discoverycited by 0
  8. 2024Disentangled Representations for Causal Cognitioncited by 0
  9. 2025Causal Policy Learning in Reinforcement Learning: Backdoor-Adjusted Soft Actor-Criticcited by 0
  10. 2026Causal Imitation Learning Under Measurement Error and Distribution Shiftcited by 0
  11. 2026Quantifying Potential Observation Missingness in Inverse Reinforcement Learningcited by 0
  12. 2026Scalable Causal Imitation Learningcited by 0
  13. 2026Learning Implicit Causal World Models from Multi-Agent Demonstrationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Causal Imitation Learning with Unobserved Confounders
  2. 2023Causal Imitability Under Context-Specific Independence Relations
  3. 2025Confounded Causal Imitation Learning with Instrumental Variables
  4. 2021Feedback in Imitation Learning: The Three Regimes of Covariate Shift
  5. 2023Invariant Causal Imitation Learning for Generalizable Policies
  6. 2025Improving Generalization Ability of Robotic Imitation Learning by Resolving Causal Confusion in Observations
  7. 2021Fully General Online Imitation Learning
  8. 2024Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning