RuDi: Explaining Behavior Sequence Models by Automatic Statistics Generation and Rule Distillation
This work addresses the need for explainable AI in sensitive applications like risk scoring, offering a practical solution to bridge the gap between effective but opaque models and interpretable rule systems.
The paper tackles the problem of black-box risk scoring models lacking explainability by proposing RuDi, a two-stage method that distills knowledge from deep learning models into transparent rule-based systems, achieving competitive performance on multiple real-world datasets.
Risk scoring systems have been widely deployed in many applications, which assign risk scores to users according to their behavior sequences. Though many deep learning methods with sophisticated designs have achieved promising results, the black-box nature hinders their applications due to fairness, explainability, and compliance consideration. Rule-based systems are considered reliable in these sensitive scenarios. However, building a rule system is labor-intensive. Experts need to find informative statistics from user behavior sequences, design rules based on statistics and assign weights to each rule. In this paper, we bridge the gap between effective but black-box models and transparent rule models. We propose a two-stage method, RuDi, that distills the knowledge of black-box teacher models into rule-based student models. We design a Monte Carlo tree search-based statistics generation method that can provide a set of informative statistics in the first stage. Then statistics are composed into logical rules with our proposed neural logical networks by mimicking the outputs of teacher models. We evaluate RuDi on three real-world public datasets and an industrial dataset to demonstrate its effectiveness.