# Causal Imitability Under Context-Specific Independence Relations

arXiv: https://arxiv.org/abs/2306.00585 · HTML: https://mlsanity.com/paper/2306.00585/

## Key facts

- Cited by 6 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 17 of its 45 references that are in the map.
- First posted on arXiv 1 Jun 2023.

## Placed in

- Offline RL and imitation learning from data / [Imitation learning methods](https://mlsanity.com/offlinerl/imitation-learning-methods/)

## Builds on (most cited first, 17 of 17)

- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst](https://mlsanity.com/paper/1812.03079/) (2018)
- [Exploring the Limitations of Behavior Cloning for Autonomous Driving](https://mlsanity.com/paper/1904.08980/) (2019)
- [Reward learning from human preferences and demonstrations in Atari](https://mlsanity.com/paper/1811.06521/) (2018)
- [Causal Confusion in Imitation Learning](https://mlsanity.com/paper/1905.11979/) (2019)
- [Imitating Driver Behavior with Generative Adversarial Networks](https://mlsanity.com/paper/1701.06699/) (2017)
- [Context-Specific Independence in Bayesian Networks](https://mlsanity.com/paper/1302.3562/) (2013)
- [Causal Imitation Learning with Unobserved Confounders](https://mlsanity.com/paper/2208.06267/) (2022)
- [Recursive Causal Structure Learning in the Presence of Latent Variables and Selection Bias](https://mlsanity.com/paper/2110.12036/) (2021)
- [Causal Imitation Learning under Temporally Correlated Noise](https://mlsanity.com/paper/2202.01312/) (2022)
- [Identifying Causal Effects via Context-specific Independence Relations](https://mlsanity.com/paper/2009.09768/) (2020)
- [Labeled Directed Acyclic Graphs: a generalization of context-specific independence in directed graphical models](https://mlsanity.com/paper/1310.1187/) (2013)
- [Causal Transfer for Imitation Learning and Decision Making under Sensor-shift](https://mlsanity.com/paper/2003.00806/) (2020)
- [A New Perspective on Learning Context-Specific Independence](https://mlsanity.com/paper/2006.06896/) (2020)
- [Causal Effect Identification with Context-specific Independence Relations of Control Variables](https://mlsanity.com/paper/2110.12064/) (2021)
- [An Algorithm for Finding Minimum d-Separating Sets in Belief Networks](https://mlsanity.com/paper/1302.3549/) (2013)
- [Causal Effect Identification in Uncertain Causal Networks](https://mlsanity.com/paper/2208.04627/) (2022)

## Built on by (most cited first, 6 of 6)

- [Fine-Grained Causal Dynamics Learning with Quantization for Improving Robustness in Reinforcement Learning](https://mlsanity.com/paper/2406.03234/) (2024)
- [Confounded Budgeted Causal Bandits](https://mlsanity.com/paper/2401.07578/) (2024)
- [Curious Causality-Seeking Agents Learn Meta Causal World](https://mlsanity.com/paper/2506.23068/) (2025)
- [Fast Proxy Experiment Design for Causal Effect Identification](https://mlsanity.com/paper/2407.05330/) (2024)
- [Causal Effect Identification in a Sub-Population with Latent Variables](https://mlsanity.com/paper/2405.14547/) (2024)
- [Causal Imitation Learning under Expert-Observable and Expert-Unobservable Confounding](https://mlsanity.com/paper/2502.07656/) (2025)

## Abstract (arXiv)

Drawbacks of ignoring the causal mechanisms when performing imitation learning have recently been acknowledged. Several approaches both to assess the feasibility of imitation and to circumvent causal confounding and causal misspecifications have been proposed in the literature. However, the potential benefits of the incorporation of additional information about the underlying causal structure are left unexplored. An example of such overlooked information is context-specific independence (CSI), i.e., independence that holds only in certain contexts. We consider the problem of causal imitation learning when CSI relations are known. We prove that the decision problem pertaining to the feasibility of imitation in this setting is NP-hard. Further, we provide a necessary graphical criterion for imitation learning under CSI and show that under a structural assumption, this criterion is also sufficient. Finally, we propose a sound algorithmic approach for causal imitation learning which takes both CSI relations and data into account.