diff --git a/docs/source/explainable_sir.ipynb b/docs/source/explainable_sir.ipynb index 7b0c0ff1..f79cc761 100644 --- a/docs/source/explainable_sir.ipynb +++ b/docs/source/explainable_sir.ipynb @@ -1744,7 +1744,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "`SearchForExplanation` can be used in any `ChiRho` causal model to investigate the causal role of variables of interest in outome variables having the values that they do in particular runs. This is achieved by transforming the causal model into one that allows us to investigate multiple causal settings with \"necessity\" and \"sufficiency\" interventions and their impact on the outcome variable across different possible worlds. Looking back at the heatmap, \"Counterfactual heatmap of overshoot under necessity and sufficiency interventions\" with the accompanying explanation illustrates how paying attention to both conjoining sufficiency and necessity and to searching through actual contexts to be fixed allows us to break the *prima facie* symmetries which makes the usual but-for analysis blind to such a fine-grained analysis. Moreover, the obtained samples can be used to answer more specific causal queries if need be." + "`SearchForExplanation` can be used in any `ChiRho` causal model to investigate the causal role of variables of interest in outome variables having the values that they do in particular runs. This is achieved by transforming the causal model into one that allows us to investigate multiple causal settings with \"necessity\" and \"sufficiency\" interventions and their impact on the outcome variable across different possible worlds. Looking back at the heatmap, \"Counterfactual heatmap of overshoot under necessity and sufficiency interventions\" with the accompanying explanation illustrates how paying attention to both conjoining sufficiency and necessity and to searching through actual contexts to be fixed allows us to break the *prima facie* symmetries which makes the usual but-for analysis blind to such a fine-grained analysis. Moreover, the obtained samples can be used to answer more specific causal queries if need be.\n", + "\n", + "A practical policy-making implication is that developing causal models for problems at hand and employing a strategy analogous to the one discussed in this notebook is a promising strategy for a fine-grained and probabilistically aware responsibility attribution." ] } ],