Decision making, symmetry and structure: Justifying causal interventions
We can use structural causal models (SCMs) to help us evaluate the consequences of actions given data. SCMs identify actions with structural interventions. A careful decision maker may wonder whether this identification is justified. We seek such a justification. We begin with decision models, which...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2025-01-01
|
| Series: | Journal of Causal Inference |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/jci-2023-0001 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832593652647460864 |
|---|---|
| author | Johnston David O. Ong Cheng Soon Williamson Robert C. |
| author_facet | Johnston David O. Ong Cheng Soon Williamson Robert C. |
| author_sort | Johnston David O. |
| collection | DOAJ |
| description | We can use structural causal models (SCMs) to help us evaluate the consequences of actions given data. SCMs identify actions with structural interventions. A careful decision maker may wonder whether this identification is justified. We seek such a justification. We begin with decision models, which map actions to distributions over outcomes but avoid additional causal assumptions. We then examine assumptions that could justify causal interventions, with a focus on symmetry. First, we introduce conditionally independent and identical responses (CIIR), a generalisation of the IID assumption to decision models. CIIR justifies identifying actions with interventions, but is often an implausible assumption. We consider an alternative: precedent is the assumption that “what I can do has been done before, and its consequences observed,” and is generally more plausible than CIIR. We show that precedent together with independence of causal mechanisms (ICM) and an observed conditional independence can justify identifying actions with causal interventions. ICM has been proposed as an alternative foundation for causal modelling, but this work suggests that it may in fact justify the interventional interpretation of causal models. |
| format | Article |
| id | doaj-art-ecb50250201c44fe8ba913a0573d0a88 |
| institution | Kabale University |
| issn | 2193-3685 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Journal of Causal Inference |
| spelling | doaj-art-ecb50250201c44fe8ba913a0573d0a882025-01-20T11:08:51ZengDe GruyterJournal of Causal Inference2193-36852025-01-01131S8S1410.1515/jci-2023-0001Decision making, symmetry and structure: Justifying causal interventionsJohnston David O.0Ong Cheng Soon1Williamson Robert C.2Interpretability Team, Eleuther AI, Berkeley, USAMachine Learning Research Group, Data61, Canberra, AustraliaFaculty of Mathematics and Natural Sciences, Universität Tübingen and Tübingen AI center, Tübingen, GermanyWe can use structural causal models (SCMs) to help us evaluate the consequences of actions given data. SCMs identify actions with structural interventions. A careful decision maker may wonder whether this identification is justified. We seek such a justification. We begin with decision models, which map actions to distributions over outcomes but avoid additional causal assumptions. We then examine assumptions that could justify causal interventions, with a focus on symmetry. First, we introduce conditionally independent and identical responses (CIIR), a generalisation of the IID assumption to decision models. CIIR justifies identifying actions with interventions, but is often an implausible assumption. We consider an alternative: precedent is the assumption that “what I can do has been done before, and its consequences observed,” and is generally more plausible than CIIR. We show that precedent together with independence of causal mechanisms (ICM) and an observed conditional independence can justify identifying actions with causal interventions. ICM has been proposed as an alternative foundation for causal modelling, but this work suggests that it may in fact justify the interventional interpretation of causal models.https://doi.org/10.1515/jci-2023-0001causal inferencedecision theory62d2062a0168t3760a99 |
| spellingShingle | Johnston David O. Ong Cheng Soon Williamson Robert C. Decision making, symmetry and structure: Justifying causal interventions Journal of Causal Inference causal inference decision theory 62d20 62a01 68t37 60a99 |
| title | Decision making, symmetry and structure: Justifying causal interventions |
| title_full | Decision making, symmetry and structure: Justifying causal interventions |
| title_fullStr | Decision making, symmetry and structure: Justifying causal interventions |
| title_full_unstemmed | Decision making, symmetry and structure: Justifying causal interventions |
| title_short | Decision making, symmetry and structure: Justifying causal interventions |
| title_sort | decision making symmetry and structure justifying causal interventions |
| topic | causal inference decision theory 62d20 62a01 68t37 60a99 |
| url | https://doi.org/10.1515/jci-2023-0001 |
| work_keys_str_mv | AT johnstondavido decisionmakingsymmetryandstructurejustifyingcausalinterventions AT ongchengsoon decisionmakingsymmetryandstructurejustifyingcausalinterventions AT williamsonrobertc decisionmakingsymmetryandstructurejustifyingcausalinterventions |