Consequence-aware Sequential Counterfactual Generation

TL;DR

This paper introduces a model-agnostic, multi-objective genetic algorithm for sequential counterfactual generation that accounts for action consequences, producing less costly, diverse, and efficient solutions.

Contribution

It presents a novel method that considers action sequences and their consequences, improving over existing approaches in counterfactual generation.

Findings

Generates less costly counterfactuals

More efficient than state-of-the-art methods

Provides diverse solution options

Abstract

Counterfactuals have become a popular technique nowadays for interacting with black-box machine learning models and understanding how to change a particular instance to obtain a desired outcome from the model. However, most existing approaches assume instant materialization of these changes, ignoring that they may require effort and a specific order of application. Recently, methods have been proposed that also consider the order in which actions are applied, leading to the so-called sequential counterfactual generation problem. In this work, we propose a model-agnostic method for sequential counterfactual generation. We formulate the task as a multi-objective optimization problem and present a genetic algorithm approach to find optimal sequences of actions leading to the counterfactuals. Our cost model considers not only the direct effect of an action, but also its consequences. Experimental results show that compared to state-of-the-art, our approach generates less costly solutions, is more efficient and provides the user with a diverse set of solutions to choose from.