Inverse Classification with Limited Budget and Maximum Number of Perturbed Samples
This work addresses the need for interpretability and actionable insights in business and healthcare applications, such as diet recommendations for diabetes patients, by providing a practical solution for adjusting sample inputs to achieve desired classifier predictions, though it is incremental as it builds on existing inverse classification methods with budget considerations.
The authors tackled the problem of inverse classification under budget constraints by proposing a framework that maximizes the number of perturbed samples while adhering to per-feature budget limits and ensuring favorable classification outcomes, with their stochastic process-based algorithms showing excellent performance and scalability in experiments.
Most recent machine learning research focuses on developing new classifiers for the sake of improving classification accuracy. With many well-performing state-of-the-art classifiers available, there is a growing need for understanding interpretability of a classifier necessitated by practical purposes such as to find the best diet recommendation for a diabetes patient. Inverse classification is a post modeling process to find changes in input features of samples to alter the initially predicted class. It is useful in many business applications to determine how to adjust a sample input data such that the classifier predicts it to be in a desired class. In real world applications, a budget on perturbations of samples corresponding to customers or patients is usually considered, and in this setting, the number of successfully perturbed samples is key to increase benefits. In this study, we propose a new framework to solve inverse classification that maximizes the number of perturbed samples subject to a per-feature-budget limits and favorable classification classes of the perturbed samples. We design algorithms to solve this optimization problem based on gradient methods, stochastic processes, Lagrangian relaxations, and the Gumbel trick. In experiments, we find that our algorithms based on stochastic processes exhibit an excellent performance in different budget settings and they scale well.