Constrained Generalized Additive 2 Model With Consideration of High-Order Interactions

In recent years, machine learning and AI have been introduced in many industrial fields. In fields such as finance, medicine, and autonomous driving, where the inference results of a model may have serious consequences, high interpretability, as well as prediction accuracy is required. In this study, we propose Constrained Generalized Additive 2 Model plus (CGA ² M+), which is based on the Generalized Additive 2 Model (GA ² M) and differs from it in two major points. The first is the introduction of monotonicity. Enforcing monotonicity constraints on some features using an analyst's knowledge, is expected to improve not only interpretability but also generalization performance. The second is the introduction of a higher-order term: given that GA2M considers only up to second-order interactions, we introduce a higher-order term, which can capture higher-order interactions and improve prediction performance. In this process, the higher-order term does not reduce interpretability by applying our training method. Numerical experiments showed that the proposed model has high prediction accuracy and interpretability. Furthermore, we confirmed that generalization performance is improved by imposing monotonicity constraints.