Extending Structural Causal Models for Autonomous Vehicles to Simplify Temporal System Construction & Enable Dynamic Interactions Between Agents
Oxford Robotics Institute, University of Oxford（牛津大学牛津机器人研究所）
Rhys Howard, Lars Kunze
备注：
30 Pages = 13 Pages (Main Content) + 4 Pages (References) + 13 Pages (Appendix), 15 Figures = 5 Figures (Main Content) + 10 (Appendix), To be published in the Proceedings of the 2025 Causal Learning and Reasoning Conference, Update upload of accepted paper version
In this work we aim to bridge the divide between autonomous vehicles and causal reasoning. Autonomous vehicles have come to increasingly interact with human drivers, and in many cases may pose risks to the physical or mental well-being of those they interact with. Meanwhile causal models, despite their inherent transparency and ability to offer contrastive explanations, have found limited usage within such systems. As such, we first identify the challenges that have limited the integration of structural causal models within autonomous vehicles. We then introduce a number of theoretical extensions to the structural causal model formalism in order to tackle these challenges. This augments these models to possess greater levels of modularisation and encapsulation, as well presenting temporal causal model representation with constant space complexity. We also prove through the extensions we have introduced that dynamically mutable sets (e.g. varying numbers of autonomous vehicles across time) can be used within a structural causal model while maintaining a relaxed form of causal stationarity. Finally we discuss the application of the extensions in the context of the autonomous vehicle and service robotics domain along with potential directions for future work.
Do Not Marginalize Mechanisms, Rather Consolidate!
Technical University of Darmstadt（达姆施塔特工业大学）；Eindhoven University of Technology（埃因霍温理工大学）；Hessian Center for AI (hessian.AI)（黑森人工智能中心（hessian.AI））；German Research Center for AI (DFKI)（德国人工智能研究中心（DFKI））
Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Series
Department of Electrical and Computer Engineering, Duke University, Durham, NC 15213, USA（电气与计算机工程系，杜克大学，达勒姆，NC 15213，美国）；Departments of Civil and Environmental Engineering（土木与环境工程系；生物统计学与生物信息学系；和计算机科学系，杜克大学，达勒姆，NC 15213，美国）；Biostatistics and Bioinformatics；and Computer Science, Duke University, Durham, NC 15213, USA
