Estimating counterfactual treatment outcomes over time in multi-vehicle simulation

Keisuke Fujii, Koh Takeuchi, Atsushi Kuribayashi, Naoya Takeishi, Yoshinobu Kawahara, Kazuya Takeda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Evaluation of intervention in a multi-agent system, e.g., when humans should intervene in autonomous driving systems, is challenging in various engineering and scientific fields. Estimating the individual treatment effect (ITE) using counterfactual long-term prediction is practical to evaluate such interventions. However, most of the conventional frameworks did not consider the time-varying complex structure of multi-agent relationships and covariate counterfactual prediction. Here we propose an interpretable, counterfactual recurrent network in multi-agent systems to estimate the effect of the intervention. Our model leverages graph variational recurrent neural networks and theory-based computation with domain knowledge for the ITE estimation framework based on long-term prediction of multi-agent covariates and outcomes, which can confirm the circumstances under which the intervention is effective. On simulated models of an automated vehicle with time-varying confounders, we show that our methods achieved lower estimation errors in counterfactual covariates.

Original languageEnglish
Title of host publication30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL GIS 2022
EditorsMatthias Renz, Mohamed Sarwat, Mario A. Nascimento, Shashi Shekhar, Xing Xie
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450395298
DOIs
Publication statusPublished - Nov 1 2022
Event30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, SIGSPATIAL GIS 2022 - Seattle, United States
Duration: Nov 1 2022Nov 4 2022

Publication series

NameGIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems

Conference

Conference30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, SIGSPATIAL GIS 2022
Country/TerritoryUnited States
CitySeattle
Period11/1/2211/4/22

All Science Journal Classification (ASJC) codes

  • Earth-Surface Processes
  • Computer Science Applications
  • Modelling and Simulation
  • Computer Graphics and Computer-Aided Design
  • Information Systems

Fingerprint

Dive into the research topics of 'Estimating counterfactual treatment outcomes over time in multi-vehicle simulation'. Together they form a unique fingerprint.

Cite this