Causes of Effects: Learning individual responses from population data
This work addresses the crucial problem of individualization in decision-making fields like medicine and AI, but it is incremental as it builds on and expands existing research on causal bounds.
The paper tackles the problem of estimating individual causal effects, such as the probability that a treatment benefits a specific person, from population data, which is inherently challenging due to counterfactual indeterminacy. It shows that using structural causal models can yield significantly narrower bounds on these probabilities compared to existing tight bounds, enabling applications in explainable AI, legal responsibility, and personalized medicine.
The problem of individualization is recognized as crucial in almost every field. Identifying causes of effects in specific events is likewise essential for accurate decision making. However, such estimates invoke counterfactual relationships, and are therefore indeterminable from population data. For example, the probability of benefiting from a treatment concerns an individual having a favorable outcome if treated and an unfavorable outcome if untreated. Experiments conditioning on fine-grained features are fundamentally inadequate because we can't test both possibilities for an individual. Tian and Pearl provided bounds on this and other probabilities of causation using a combination of experimental and observational data. Even though those bounds were proven tight, narrower bounds, sometimes significantly so, can be achieved when structural information is available in the form of a causal model. This has the power to solve central problems, such as explainable AI, legal responsibility, and personalized medicine, all of which demand counterfactual logic. We analyze and expand on existing research by applying bounds to the probability of necessity and sufficiency (PNS) along with graphical criteria and practical applications.