Detecting causal relationships in simulation models using intervention-based counterfactual analysis
- Submitting institution
-
King's College London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 111092257
- Type
- D - Journal article
- DOI
-
10.1145/3322123
- Title of journal
- ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
- Article number
- 47
- First page
- 1
- Volume
- 10
- Issue
- 5
- ISSN
- 2157-6904
- Open access status
- Compliant
- Month of publication
- September
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper solved the substantial long-running challenge of reliably detecting token causation in data-scarce emergent phenomena through analysis of simulation traces. The approach does not require knowledge of the processes comprising the analysed phenomenon, so broadening the applicability and value of analysis compared to prior state of the art. Due to model complexity, agent-based models are typically analysed through statistical sampling of results. This paper provides a novel formal model-checking approach that does not require such sampling. The approach was applied to an epidemiological model, showing that the root source of a given agent infection could be identified.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -