A compositional approach to causality
From Semantic Portal Wiki
{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}
| Edit |
Reference: {{#vardefine:pagename|a compositional approach to causality }}
- [[]]
bibtex
{{#vardefine:pagename|A compositional approach to causality }}{{#vardefine:key| }}
abstract: Inferring causality from equation models characterizing engineeringdomains is important towards predicting and diagnosing systembehavior. Most previous attempts in this direction have failed torecognize the key differences between equations which model physicalphenomena and those that just express rationality or numericalconveniences of the designer. These different types of equations beardifferent causal implications among the model parameters theyrelate. We show how unstructured and ad hoc formulations of equationmodels for apparent numerical conveniences are lossy in the causalinformation encoding and justify the use of CML as a model formulationparadigm which retains these causal structures among model parametersby clearly separating equations corresponding to phenomena andrationality. We provide an algorithm to infer causality from theactive model fragments by using the notion of PreCondition graphs.
download:
- paper:
- slides:
| Abstract | Inferring causality from equation models c … Inferring causality from equation models characterizing engineeringdomains is important towards predicting and diagnosing systembehavior. Most previous attempts in this direction have failed torecognize the key differences between equations which model physicalphenomena and those that just express rationality or numericalconveniences of the designer. These different types of equations beardifferent causal implications among the model parameters theyrelate. We show how unstructured and ad hoc formulations of equationmodels for apparent numerical conveniences are lossy in the causalinformation encoding and justify the use of CML as a model formulationparadigm which retains these causal structures among model parametersby clearly separating equations corresponding to phenomena andrationality. We provide an algorithm to infer causality from theactive model fragments by using the notion of PreCondition graphs. y using the notion of PreCondition graphs. |
| Address | Austin, Texas + |
| Author | T. K. Satish Kumar + |
| Bibtype | inproceedings + |
| Booktitle | Proceedings of the Fourth International Symposium on Abstraction, Reformulation and Approximation (SARA2000) + |
| Key | KSL-00-04 + |
| Tag | Computer science + |
| Title | A Compositional Approach to Causality + |
| Tr id | KSL-00-04 + |
| Year | 2000 + |
