Mining Pareto-Optimal Counterfactual Antecedents With A Branch-And-Bound Model-Agnostic Algorithm

Resumo

Mining counterfactual antecedents became a valuable tool to discover knowledge and explain machine learning models. It consists of generating synthetic samples from an original sample to achieve the desired outcome in a machine learning model thus helping to understand the prediction. An insightful methodology would explore a broader set of counterfactual antecedents to reveal multiple possibilities while operating on any classifier. Thus, we create a treebased search that requires monotonicity from the objective functions (a.k.a. cost functions); it allows pruning branches that will not improve the objective functions. Since monotonicity is only required for the objective function, this method can be used for any family of classifiers (e.g., linear models, neural networks, decision trees). However, additional classifier properties speed up the tree-search when it foresees branches that will not result in feasible actions. Moreover, the proposed optimization generates a diverse set of Pareto-optimal counterfactual antecedents by relying on multi-objective concepts. The results show an algorithm with working guarantees that enumerates a wide range of counterfactual antecedents. It helps the decision-maker understand the machine learning decision and finds alternatives to achieve the desired outcome. The user can inspect these multiple counterfactual antecedents to find the most suitable one and have a broader understanding of the prediction.

Ficha do documento

Tipo

Outro

Ano

2021

Instituição

Fundação Getulio Vargas

Fonte

Repositório da FGV

Idioma

Inglês

Acesso

Não informado

Identificador

oai:repositorio.fgv.br:10438/33951

Temas

Economia

Palavras-chave

CounterfactualInterpretabilityMachine learningKnowledge miningMulti-objective optimizationCrime dataEconomiaAprendizado do computador