CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning

Daniel McDuff, Yale Song, Jiyoung Lee, Vibhav Vineet, Sai Vemprala, Nicholas Alexander Gyde, Hadi Salman, Shuang Ma, Kwanghoon Sohn, Ashish Kapoor
Proceedings of the First Conference on Causal Learning and Reasoning, PMLR 177:559-575, 2022.

Abstract

The ability to perform causal and counterfactual reasoning are central properties of human intelligence. Decision-making systems that can perform these types of reasoning have the potential to be more generalizable and interpretable. Simulations have helped advance the state-of-the-art in this domain, by providing the ability to systematically vary parameters (e.g., confounders) and generate examples of the outcomes in the case of counterfactual scenarios. However, simulating complex temporal causal events in multi-agent scenarios, such as those that exist in driving and vehicle navigation, is challenging. To help address this, we present a high-fidelity simulation environment that is designed for developing algorithms for causal discovery and counterfactual reasoning in the safety-critical context. A core component of our work is to introduce agency, such that it is simple to define and create complex scenarios using high-level definitions. The vehicles then operate with agency to complete these objectives, meaning low-level behaviors need only be controlled if necessary. We perform experiments with three state-of-the-art methods to create baselines and highlight the affordances of this environment. Finally, we highlight challenges and opportunities for future work.

Cite this Paper

BibTeX
@InProceedings{pmlr-v177-mcduff22a, title = {CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning}, author = {McDuff, Daniel and Song, Yale and Lee, Jiyoung and Vineet, Vibhav and Vemprala, Sai and Gyde, Nicholas Alexander and Salman, Hadi and Ma, Shuang and Sohn, Kwanghoon and Kapoor, Ashish}, booktitle = {Proceedings of the First Conference on Causal Learning and Reasoning}, pages = {559--575}, year = {2022}, editor = {Schölkopf, Bernhard and Uhler, Caroline and Zhang, Kun}, volume = {177}, series = {Proceedings of Machine Learning Research}, month = {11--13 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v177/mcduff22a/mcduff22a.pdf}, url = {https://proceedings.mlr.press/v177/mcduff22a.html}, abstract = {The ability to perform causal and counterfactual reasoning are central properties of human intelligence. Decision-making systems that can perform these types of reasoning have the potential to be more generalizable and interpretable. Simulations have helped advance the state-of-the-art in this domain, by providing the ability to systematically vary parameters (e.g., confounders) and generate examples of the outcomes in the case of counterfactual scenarios. However, simulating complex temporal causal events in multi-agent scenarios, such as those that exist in driving and vehicle navigation, is challenging. To help address this, we present a high-fidelity simulation environment that is designed for developing algorithms for causal discovery and counterfactual reasoning in the safety-critical context. A core component of our work is to introduce agency, such that it is simple to define and create complex scenarios using high-level definitions. The vehicles then operate with agency to complete these objectives, meaning low-level behaviors need only be controlled if necessary. We perform experiments with three state-of-the-art methods to create baselines and highlight the affordances of this environment. Finally, we highlight challenges and opportunities for future work.} }
Endnote
%0 Conference Paper %T CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning %A Daniel McDuff %A Yale Song %A Jiyoung Lee %A Vibhav Vineet %A Sai Vemprala %A Nicholas Alexander Gyde %A Hadi Salman %A Shuang Ma %A Kwanghoon Sohn %A Ashish Kapoor %B Proceedings of the First Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2022 %E Bernhard Schölkopf %E Caroline Uhler %E Kun Zhang %F pmlr-v177-mcduff22a %I PMLR %P 559--575 %U https://proceedings.mlr.press/v177/mcduff22a.html %V 177 %X The ability to perform causal and counterfactual reasoning are central properties of human intelligence. Decision-making systems that can perform these types of reasoning have the potential to be more generalizable and interpretable. Simulations have helped advance the state-of-the-art in this domain, by providing the ability to systematically vary parameters (e.g., confounders) and generate examples of the outcomes in the case of counterfactual scenarios. However, simulating complex temporal causal events in multi-agent scenarios, such as those that exist in driving and vehicle navigation, is challenging. To help address this, we present a high-fidelity simulation environment that is designed for developing algorithms for causal discovery and counterfactual reasoning in the safety-critical context. A core component of our work is to introduce agency, such that it is simple to define and create complex scenarios using high-level definitions. The vehicles then operate with agency to complete these objectives, meaning low-level behaviors need only be controlled if necessary. We perform experiments with three state-of-the-art methods to create baselines and highlight the affordances of this environment. Finally, we highlight challenges and opportunities for future work.
APA
McDuff, D., Song, Y., Lee, J., Vineet, V., Vemprala, S., Gyde, N.A., Salman, H., Ma, S., Sohn, K. & Kapoor, A.. (2022). CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning. Proceedings of the First Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 177:559-575 Available from https://proceedings.mlr.press/v177/mcduff22a.html.

Related Material