continuous FeAR

Quantifying Causal Responsibility in Continuous Spatial Interactions via Feasible Action Space Reduction

Github repository for :

"Quantifying Causal Responsibility in Continuous Spatial Interactions via Feasible Action Space Reduction" by Ashwin George, Luciano Cavalcante Siebert, David A. Abbink and Arkady Zgonnikov. https://arxiv.org/abs/2505.17739

We propose Feasible Action-Space Reduction (FeAR) as a metric for causal responsibility in multi-agent spatial interactions with continuous actions.

---

The code in this repository is based on the code for our earlier publication which can be found at: https://github.com/DAI-Lab-HERALD/FeAR

---

---

---

Contents

To view the results for the case studies in the paper

• Download the .pkl files for results from: https://osf.io/8mtnf/ and place them in \FeAR_Results\.
• Run cFeAR_Paper_Read_and_Plot_Results.ipynb to generate plots.

To run simulations

• Run ContinuousFeAR_Sims_Scenario.ipynb
• Run ContinuousFeAR_Sims_MdR_Confusion.ipynb to run simulations with different joint actions as MdRs
• These use scenarios defined in Scenarios4cFeAR.json

To generate other illustrations in the paper

• Run Illustrations.ipynb

Supplementary Files and Data

• https://osf.io/8mtnf/

Authors

• Ashwin George - PhD student @ TU Delft

Funding and support

• TU Delft AI Initiative

Reference

Copyright notice

Technische Universiteit Delft hereby disclaims all copyright interest in the program “FeAR” (Feasible Action-Space Reduction) written by Ashwin George. Fred van Keulen, Dean of Mechanical Engineering

© 2025, Ashwin George, HERALD Lab, TU Delft